From 703486b45e99dbb51ba47a6b1524099c0214a76a Mon Sep 17 00:00:00 2001
From: gohabereg <gohabereg@gmail.com>
Date: Fri, 17 Jul 2026 20:32:26 +0100
Subject: [PATCH 1/5] Init openspec & generate specs on existing codebase

---
 .claude/commands/opsx/apply.md                | 156 +++++
 .claude/commands/opsx/archive.md              | 161 +++++
 .claude/commands/opsx/explore.md              | 175 ++++++
 .claude/commands/opsx/onboard.md              | 551 +++++++++++++++++
 .claude/commands/opsx/propose.md              | 110 ++++
 .claude/commands/opsx/sync.md                 | 144 +++++
 .claude/commands/opsx/update.md               |  82 +++
 .claude/commands/opsx/verify.md               | 168 ++++++
 .claude/skills/openspec-apply-change/SKILL.md | 160 +++++
 .../skills/openspec-archive-change/SKILL.md   | 118 ++++
 .claude/skills/openspec-explore/SKILL.md      | 290 +++++++++
 .../skills/openspec-generate-specs/SKILL.md   | 146 +++++
 .claude/skills/openspec-onboard/SKILL.md      | 555 ++++++++++++++++++
 .claude/skills/openspec-propose/SKILL.md      | 114 ++++
 .claude/skills/openspec-sync-specs/SKILL.md   | 148 +++++
 .../skills/openspec-update-change/SKILL.md    |  86 +++
 .../skills/openspec-verify-change/SKILL.md    | 172 ++++++
 .github/prompts/opsx-apply.prompt.md          | 152 +++++
 .github/prompts/opsx-archive.prompt.md        | 157 +++++
 .github/prompts/opsx-explore.prompt.md        | 171 ++++++
 .github/prompts/opsx-onboard.prompt.md        | 547 +++++++++++++++++
 .github/prompts/opsx-propose.prompt.md        | 106 ++++
 .github/prompts/opsx-sync.prompt.md           | 140 +++++
 .github/prompts/opsx-update.prompt.md         |  78 +++
 .github/prompts/opsx-verify.prompt.md         | 164 ++++++
 .github/skills/openspec-apply-change/SKILL.md | 160 +++++
 .../skills/openspec-archive-change/SKILL.md   | 118 ++++
 .github/skills/openspec-explore/SKILL.md      | 290 +++++++++
 .github/skills/openspec-onboard/SKILL.md      | 555 ++++++++++++++++++
 .github/skills/openspec-propose/SKILL.md      | 114 ++++
 .github/skills/openspec-sync-specs/SKILL.md   | 148 +++++
 .../skills/openspec-update-change/SKILL.md    |  86 +++
 .../skills/openspec-verify-change/SKILL.md    | 172 ++++++
 .gitignore                                    |   2 +-
 .opencode/commands/opsx-apply.md              | 152 +++++
 .opencode/commands/opsx-archive.md            | 157 +++++
 .opencode/commands/opsx-explore.md            | 171 ++++++
 .opencode/commands/opsx-onboard.md            | 547 +++++++++++++++++
 .opencode/commands/opsx-propose.md            | 106 ++++
 .opencode/commands/opsx-sync.md               | 140 +++++
 .opencode/commands/opsx-update.md             |  78 +++
 .opencode/commands/opsx-verify.md             | 164 ++++++
 .../skills/openspec-apply-change/SKILL.md     | 160 +++++
 .../skills/openspec-archive-change/SKILL.md   | 118 ++++
 .opencode/skills/openspec-explore/SKILL.md    | 290 +++++++++
 .opencode/skills/openspec-onboard/SKILL.md    | 555 ++++++++++++++++++
 .opencode/skills/openspec-propose/SKILL.md    | 114 ++++
 .opencode/skills/openspec-sync-specs/SKILL.md | 148 +++++
 .../skills/openspec-update-change/SKILL.md    |  86 +++
 .../skills/openspec-verify-change/SKILL.md    | 172 ++++++
 openspec/config.yaml                          |  20 +
 openspec/specs/architecture/spec.md           |  77 +++
 openspec/specs/collaboration-manager/spec.md  | 107 ++++
 openspec/specs/core/spec.md                   | 102 ++++
 openspec/specs/dom-adapters/spec.md           |  82 +++
 openspec/specs/model-types/spec.md            |  70 +++
 openspec/specs/model/spec.md                  | 112 ++++
 openspec/specs/ot-server/spec.md              |  97 +++
 openspec/specs/sdk/spec.md                    |  89 +++
 openspec/specs/tools/spec.md                  |  94 +++
 openspec/specs/ui/spec.md                     |  94 +++
 packages/.claude/commands/opsx/apply.md       | 156 +++++
 packages/.claude/commands/opsx/archive.md     | 161 +++++
 packages/.claude/commands/opsx/explore.md     | 175 ++++++
 packages/.claude/commands/opsx/propose.md     | 110 ++++
 packages/.claude/commands/opsx/sync.md        | 144 +++++
 packages/.claude/commands/opsx/update.md      |  82 +++
 packages/.claude/commands/opsx/verify.md      | 168 ++++++
 .../skills/openspec-apply-change/SKILL.md     | 160 +++++
 .../skills/openspec-archive-change/SKILL.md   | 118 ++++
 .../.claude/skills/openspec-explore/SKILL.md  | 290 +++++++++
 .../.claude/skills/openspec-propose/SKILL.md  | 114 ++++
 .../skills/openspec-sync-specs/SKILL.md       | 148 +++++
 .../skills/openspec-update-change/SKILL.md    |  86 +++
 .../skills/openspec-verify-change/SKILL.md    | 172 ++++++
 packages/.github/prompts/opsx-apply.prompt.md | 152 +++++
 .../.github/prompts/opsx-archive.prompt.md    | 157 +++++
 .../.github/prompts/opsx-explore.prompt.md    | 171 ++++++
 .../.github/prompts/opsx-propose.prompt.md    | 106 ++++
 packages/.github/prompts/opsx-sync.prompt.md  | 140 +++++
 .../.github/prompts/opsx-update.prompt.md     |  78 +++
 .../.github/prompts/opsx-verify.prompt.md     | 164 ++++++
 .../skills/openspec-apply-change/SKILL.md     | 160 +++++
 .../skills/openspec-archive-change/SKILL.md   | 118 ++++
 .../.github/skills/openspec-explore/SKILL.md  | 290 +++++++++
 .../.github/skills/openspec-propose/SKILL.md  | 114 ++++
 .../skills/openspec-sync-specs/SKILL.md       | 148 +++++
 .../skills/openspec-update-change/SKILL.md    |  86 +++
 .../skills/openspec-verify-change/SKILL.md    | 172 ++++++
 packages/openspec/config.yaml                 |  20 +
 90 files changed, 14757 insertions(+), 1 deletion(-)
 create mode 100644 .claude/commands/opsx/apply.md
 create mode 100644 .claude/commands/opsx/archive.md
 create mode 100644 .claude/commands/opsx/explore.md
 create mode 100644 .claude/commands/opsx/onboard.md
 create mode 100644 .claude/commands/opsx/propose.md
 create mode 100644 .claude/commands/opsx/sync.md
 create mode 100644 .claude/commands/opsx/update.md
 create mode 100644 .claude/commands/opsx/verify.md
 create mode 100644 .claude/skills/openspec-apply-change/SKILL.md
 create mode 100644 .claude/skills/openspec-archive-change/SKILL.md
 create mode 100644 .claude/skills/openspec-explore/SKILL.md
 create mode 100644 .claude/skills/openspec-generate-specs/SKILL.md
 create mode 100644 .claude/skills/openspec-onboard/SKILL.md
 create mode 100644 .claude/skills/openspec-propose/SKILL.md
 create mode 100644 .claude/skills/openspec-sync-specs/SKILL.md
 create mode 100644 .claude/skills/openspec-update-change/SKILL.md
 create mode 100644 .claude/skills/openspec-verify-change/SKILL.md
 create mode 100644 .github/prompts/opsx-apply.prompt.md
 create mode 100644 .github/prompts/opsx-archive.prompt.md
 create mode 100644 .github/prompts/opsx-explore.prompt.md
 create mode 100644 .github/prompts/opsx-onboard.prompt.md
 create mode 100644 .github/prompts/opsx-propose.prompt.md
 create mode 100644 .github/prompts/opsx-sync.prompt.md
 create mode 100644 .github/prompts/opsx-update.prompt.md
 create mode 100644 .github/prompts/opsx-verify.prompt.md
 create mode 100644 .github/skills/openspec-apply-change/SKILL.md
 create mode 100644 .github/skills/openspec-archive-change/SKILL.md
 create mode 100644 .github/skills/openspec-explore/SKILL.md
 create mode 100644 .github/skills/openspec-onboard/SKILL.md
 create mode 100644 .github/skills/openspec-propose/SKILL.md
 create mode 100644 .github/skills/openspec-sync-specs/SKILL.md
 create mode 100644 .github/skills/openspec-update-change/SKILL.md
 create mode 100644 .github/skills/openspec-verify-change/SKILL.md
 create mode 100644 .opencode/commands/opsx-apply.md
 create mode 100644 .opencode/commands/opsx-archive.md
 create mode 100644 .opencode/commands/opsx-explore.md
 create mode 100644 .opencode/commands/opsx-onboard.md
 create mode 100644 .opencode/commands/opsx-propose.md
 create mode 100644 .opencode/commands/opsx-sync.md
 create mode 100644 .opencode/commands/opsx-update.md
 create mode 100644 .opencode/commands/opsx-verify.md
 create mode 100644 .opencode/skills/openspec-apply-change/SKILL.md
 create mode 100644 .opencode/skills/openspec-archive-change/SKILL.md
 create mode 100644 .opencode/skills/openspec-explore/SKILL.md
 create mode 100644 .opencode/skills/openspec-onboard/SKILL.md
 create mode 100644 .opencode/skills/openspec-propose/SKILL.md
 create mode 100644 .opencode/skills/openspec-sync-specs/SKILL.md
 create mode 100644 .opencode/skills/openspec-update-change/SKILL.md
 create mode 100644 .opencode/skills/openspec-verify-change/SKILL.md
 create mode 100644 openspec/config.yaml
 create mode 100644 openspec/specs/architecture/spec.md
 create mode 100644 openspec/specs/collaboration-manager/spec.md
 create mode 100644 openspec/specs/core/spec.md
 create mode 100644 openspec/specs/dom-adapters/spec.md
 create mode 100644 openspec/specs/model-types/spec.md
 create mode 100644 openspec/specs/model/spec.md
 create mode 100644 openspec/specs/ot-server/spec.md
 create mode 100644 openspec/specs/sdk/spec.md
 create mode 100644 openspec/specs/tools/spec.md
 create mode 100644 openspec/specs/ui/spec.md
 create mode 100644 packages/.claude/commands/opsx/apply.md
 create mode 100644 packages/.claude/commands/opsx/archive.md
 create mode 100644 packages/.claude/commands/opsx/explore.md
 create mode 100644 packages/.claude/commands/opsx/propose.md
 create mode 100644 packages/.claude/commands/opsx/sync.md
 create mode 100644 packages/.claude/commands/opsx/update.md
 create mode 100644 packages/.claude/commands/opsx/verify.md
 create mode 100644 packages/.claude/skills/openspec-apply-change/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-archive-change/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-explore/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-propose/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-sync-specs/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-update-change/SKILL.md
 create mode 100644 packages/.claude/skills/openspec-verify-change/SKILL.md
 create mode 100644 packages/.github/prompts/opsx-apply.prompt.md
 create mode 100644 packages/.github/prompts/opsx-archive.prompt.md
 create mode 100644 packages/.github/prompts/opsx-explore.prompt.md
 create mode 100644 packages/.github/prompts/opsx-propose.prompt.md
 create mode 100644 packages/.github/prompts/opsx-sync.prompt.md
 create mode 100644 packages/.github/prompts/opsx-update.prompt.md
 create mode 100644 packages/.github/prompts/opsx-verify.prompt.md
 create mode 100644 packages/.github/skills/openspec-apply-change/SKILL.md
 create mode 100644 packages/.github/skills/openspec-archive-change/SKILL.md
 create mode 100644 packages/.github/skills/openspec-explore/SKILL.md
 create mode 100644 packages/.github/skills/openspec-propose/SKILL.md
 create mode 100644 packages/.github/skills/openspec-sync-specs/SKILL.md
 create mode 100644 packages/.github/skills/openspec-update-change/SKILL.md
 create mode 100644 packages/.github/skills/openspec-verify-change/SKILL.md
 create mode 100644 packages/openspec/config.yaml

diff --git a/.claude/commands/opsx/apply.md b/.claude/commands/opsx/apply.md
new file mode 100644
index 00000000..c6cb9b6d
--- /dev/null
+++ b/.claude/commands/opsx/apply.md
@@ -0,0 +1,156 @@
+---
+name: "OPSX: Apply"
+description: Implement tasks from an OpenSpec change (Experimental)
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! You can archive this change with `/opsx:archive`.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.claude/commands/opsx/archive.md b/.claude/commands/opsx/archive.md
new file mode 100644
index 00000000..df8a2f2d
--- /dev/null
+++ b/.claude/commands/opsx/archive.md
@@ -0,0 +1,161 @@
+---
+name: "OPSX: Archive"
+description: Archive a completed change in the experimental workflow
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, archive, experimental]
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Spec sync status (synced / sync skipped / no delta specs)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success (No Delta Specs)**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** No delta specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success With Warnings**
+
+```
+## Archive Complete (with warnings)
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** Sync skipped (user chose to skip)
+
+**Warnings:**
+- Archived with 2 incomplete artifacts
+- Archived with 3 incomplete tasks
+- Delta spec sync was skipped (user chose to skip)
+
+Review the archive if this was not intentional.
+```
+
+**Output On Error (Archive Exists)**
+
+```
+## Archive Failed
+
+**Change:** <change-name>
+**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+
+Target archive directory already exists.
+
+**Options:**
+1. Rename the existing archive
+2. Delete the existing archive if it's a duplicate
+3. Wait until a different date to archive
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.claude/commands/opsx/explore.md b/.claude/commands/opsx/explore.md
new file mode 100644
index 00000000..7558a631
--- /dev/null
+++ b/.claude/commands/opsx/explore.md
@@ -0,0 +1,175 @@
+---
+name: "OPSX: Explore"
+description: "Enter explore mode - think through ideas, investigate problems, clarify requirements"
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, explore, experimental, thinking]
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A change name: "add-dark-mode" (to explore in context of that change)
+- A comparison: "postgres vs sqlite for this"
+- Nothing (just enter explore mode)
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+If the user mentioned a specific change name, read its artifacts for context.
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.claude/commands/opsx/onboard.md b/.claude/commands/opsx/onboard.md
new file mode 100644
index 00000000..1b0adab8
--- /dev/null
+++ b/.claude/commands/opsx/onboard.md
@@ -0,0 +1,551 @@
+---
+name: "OPSX: Onboard"
+description: Guided onboarding - walk through a complete OpenSpec workflow cycle with narration
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, onboarding, tutorial, learning]
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx:propose` | Create a change and generate all artifacts |
+ | `/opsx:explore` | Think through problems before/during work  |
+ | `/opsx:apply`   | Implement tasks from a change              |
+ | `/opsx:archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx:new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx:continue` | Continue working on an existing change                   |
+ | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx:verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx:continue <name>` - Resume artifact creation
+- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx:propose <name>` | Create a change and generate all artifacts |
+ | `/opsx:explore`        | Think through problems (no code changes)   |
+ | `/opsx:apply <name>`   | Implement tasks                            |
+ | `/opsx:archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx:new <name>`      | Start a new change, step by step    |
+ | `/opsx:continue <name>` | Continue an existing change         |
+ | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx:verify <name>`   | Verify implementation               |
+
+Try `/opsx:propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.claude/commands/opsx/propose.md b/.claude/commands/opsx/propose.md
new file mode 100644
index 00000000..8d99588f
--- /dev/null
+++ b/.claude/commands/opsx/propose.md
@@ -0,0 +1,110 @@
+---
+name: "OPSX: Propose"
+description: Propose a new change - create it and generate all artifacts in one step
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
+
+**Steps**
+
+1. **If no input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` to start implementing."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.claude/commands/opsx/sync.md b/.claude/commands/opsx/sync.md
new file mode 100644
index 00000000..a3550cf4
--- /dev/null
+++ b/.claude/commands/opsx/sync.md
@@ -0,0 +1,144 @@
+---
+name: "OPSX: Sync"
+description: Sync delta specs from a change to main specs
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, specs, experimental]
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.claude/commands/opsx/update.md b/.claude/commands/opsx/update.md
new file mode 100644
index 00000000..75104f59
--- /dev/null
+++ b/.claude/commands/opsx/update.md
@@ -0,0 +1,82 @@
+---
+name: "OPSX: Update"
+description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.claude/commands/opsx/verify.md b/.claude/commands/opsx/verify.md
new file mode 100644
index 00000000..7f6cade1
--- /dev/null
+++ b/.claude/commands/opsx/verify.md
@@ -0,0 +1,168 @@
+---
+name: "OPSX: Verify"
+description: Verify implementation matches change artifacts before archiving
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, verify, experimental]
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/.claude/skills/openspec-apply-change/SKILL.md b/.claude/skills/openspec-apply-change/SKILL.md
new file mode 100644
index 00000000..1375861c
--- /dev/null
+++ b/.claude/skills/openspec-apply-change/SKILL.md
@@ -0,0 +1,160 @@
+---
+name: openspec-apply-change
+description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! Ready to archive this change.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.claude/skills/openspec-archive-change/SKILL.md b/.claude/skills/openspec-archive-change/SKILL.md
new file mode 100644
index 00000000..c0c169d1
--- /dev/null
+++ b/.claude/skills/openspec-archive-change/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: openspec-archive-change
+description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Whether specs were synced (if applicable)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
+
+All artifacts complete. All tasks complete.
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use openspec-sync-specs approach (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.claude/skills/openspec-explore/SKILL.md b/.claude/skills/openspec-explore/SKILL.md
new file mode 100644
index 00000000..771271a8
--- /dev/null
+++ b/.claude/skills/openspec-explore/SKILL.md
@@ -0,0 +1,290 @@
+---
+name: openspec-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+      ════════════════════════════════════════════
+
+      Awareness          Coordination         Sync
+          │                   │                 │
+          ▼                   ▼                 ▼
+      ┌────────┐         ┌────────┐        ┌────────┐
+      │Presence│         │Cursors │        │  CRDT  │
+      │  "3    │         │ Multi  │        │Conflict│
+      │online" │         │ select │        │  free  │
+      └────────┘         └────────┘        └────────┘
+          │                   │                 │
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     ┌─────────────────────────────────────────────┐
+     │              CURRENT AUTH FLOW              │
+     └─────────────────────────────────────────────┘
+                        │
+          ┌─────────────┼─────────────┐
+          ▼             ▼             ▼
+     ┌─────────┐  ┌─────────┐  ┌─────────┐
+     │ Google  │  │  GitHub │  │  Email  │
+     │  OAuth  │  │  OAuth  │  │  Magic  │
+     └────┬────┘  └────┬────┘  └────┬────┘
+          │             │             │
+          └─────────────┼─────────────┘
+                        ▼
+                  ┌───────────┐
+                  │  Session  │
+                  └─────┬─────┘
+                        │
+                        ▼
+                  ┌───────────┐
+                  │   Perms   │
+                  └───────────┘
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /opsx:explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads change artifacts]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the design to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     ┌─────────────────────────────────────────────────┐
+     │          CLI TOOL DATA STORAGE                  │
+     └─────────────────────────────────────────────────┘
+
+     Key constraints:
+     • No daemon running
+     • Must work offline
+     • Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ✓      needs server ✗
+     Offline      yes ✓           no ✗
+     Single file  yes ✓           no ✗
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a change proposal
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.claude/skills/openspec-generate-specs/SKILL.md b/.claude/skills/openspec-generate-specs/SKILL.md
new file mode 100644
index 00000000..fc1b6d79
--- /dev/null
+++ b/.claude/skills/openspec-generate-specs/SKILL.md
@@ -0,0 +1,146 @@
+---
+name: openspec-generate-specs
+description: Generate comprehensive OpenSpec specifications directly from the current project state. Use when the user wants to create or populate main specs by analyzing existing code, documentation, AGENTS.md, GitHub issues, and pull requests — without going through the change/proposal workflow. Ideal for bootstrapping specs on a project that already has working code but no specs yet, or for refreshing specs to match the current implementation.
+---
+
+Generate OpenSpec main specifications by analyzing the current project's code, documentation, issues, and pull requests.
+
+Unlike `openspec-propose` (which creates a change with proposal/design/tasks artifacts for **future** work), this skill writes finished specs directly to `openspec/specs/` based on **what already exists** in the project.
+
+**Input**: Optionally specify which capabilities or areas to generate specs for. If omitted, analyze the full project and determine capabilities automatically.
+
+**Steps**
+
+1. **Gather project context**
+
+   Collect information from all available sources. Prioritize breadth over depth — skim first, then deep-dive into relevant areas.
+
+   a. **Project documentation**:
+      - Read `AGENTS.md` (or `CONTRIBUTING.md`, `README.md`) for architecture overview, conventions, and design decisions
+      - Read `docs/` directory for feature specs, design docs, BDD features (`.feature` files)
+      - Read `openspec/config.yaml` for project context and rules
+
+   b. **Codebase structure**:
+      - Scan `src/` directory tree to understand module organization
+      - Read key files (entry points, type definitions, core modules) to understand capabilities
+      - Identify the main functional areas / capabilities the system provides
+
+   c. **GitHub issues and PRs** (if available):
+      - Use GitHub MCP tools to fetch recent open/closed issues for requirements context
+      - Use GitHub MCP tools to fetch recent merged PRs for implemented features
+      - Extract requirement-like information (feature requests, bug fixes, design decisions)
+
+   d. **Existing specs** (if any):
+      - Read `openspec/specs/` to understand what's already documented
+      - Identify gaps between existing specs and actual implementation
+
+2. **Identify capabilities**
+
+   Based on the gathered context, determine the project's main capability areas. Each capability becomes a spec file at `openspec/specs/<capability>/spec.md`.
+
+   Use the **AskUserQuestion tool** to confirm the identified capabilities before proceeding:
+   > "I've identified these capabilities from the project. Which should I generate specs for?"
+
+   Present the list with brief descriptions. Let the user select all or a subset.
+
+   **Naming convention**: Use kebab-case for capability names (e.g., `memory-system`, `platform-abstraction`, `acp-integration`).
+
+3. **Check existing specs**
+
+   For each selected capability:
+   - If `openspec/specs/<capability>/spec.md` already exists, read it
+   - Use **AskUserQuestion tool** to ask: "Spec for `<capability>` already exists. Overwrite, merge, or skip?"
+     - **Overwrite**: Replace entirely with newly generated spec
+     - **Merge**: Add missing requirements/scenarios to existing spec, preserve existing content
+     - **Skip**: Leave as-is
+
+4. **Generate specs**
+
+   For each capability, create a spec file following the OpenSpec spec format.
+
+   Use the **TodoWrite tool** to track progress through capabilities.
+
+   **Spec file structure** (`openspec/specs/<capability>/spec.md`):
+
+   ```markdown
+   # <Capability Name>
+
+   ## Purpose
+
+   <Brief description of what this capability does and why it exists.>
+
+   ## Requirements
+
+   ### Requirement: <Requirement Name>
+
+   <The system SHALL/SHOULD/MAY description using RFC 2119 language.>
+
+   #### Scenario: <Scenario Name>
+   - **GIVEN** <precondition>
+   - **WHEN** <action>
+   - **THEN** <expected outcome>
+
+   #### Scenario: <Another Scenario>
+   - **GIVEN** <precondition>
+   - **WHEN** <action>
+   - **THEN** <expected outcome>
+   ```
+
+   **Writing guidelines**:
+   - Use RFC 2119 keywords (SHALL, SHOULD, MAY) for requirement strength
+   - Derive requirements from actual code behavior, not aspirational features
+   - Include scenarios that reflect real code paths (check tests for scenario ideas)
+   - Reference source files in requirement descriptions when helpful (e.g., "Implemented in `src/core/workspace-manager.ts`")
+   - Keep requirements atomic — one concern per requirement
+   - Group related scenarios under their parent requirement
+
+5. **Write spec files**
+
+   For each capability:
+   ```bash
+   mkdir -p openspec/specs/<capability>
+   ```
+   Write the spec content to `openspec/specs/<capability>/spec.md`.
+
+   After writing each file, verify it exists and show brief progress.
+
+6. **Validate output**
+
+   Run validation if available:
+   ```bash
+   openspec validate --specs --json
+   ```
+
+   If validation fails, fix issues and re-validate.
+
+7. **Show summary**
+
+   Display what was generated:
+   ```
+   ## Specs Generated
+
+   | Capability | Requirements | Scenarios | Status |
+   |------------|-------------|-----------|--------|
+   | <name>     | N           | M         | New/Merged/Overwritten |
+
+   Total: X capabilities, Y requirements, Z scenarios
+
+   Files written to `openspec/specs/`.
+   ```
+
+**Source Priority**
+
+When information conflicts across sources, prefer in this order:
+1. Actual code behavior (ground truth)
+2. Test assertions (verified behavior)
+3. BDD feature files / design docs (intended behavior)
+4. GitHub issues/PRs (discussed behavior)
+5. README / AGENTS.md (described behavior)
+
+**Guardrails**
+- Always confirm capability list with user before generating
+- Never fabricate requirements not supported by code or documentation
+- If a capability area is ambiguous, ask for clarification
+- Preserve existing spec content when merging (only add, don't remove unless user confirms)
+- Each spec file must have at least a Purpose section and one requirement
+- Use consistent formatting across all generated spec files
diff --git a/.claude/skills/openspec-onboard/SKILL.md b/.claude/skills/openspec-onboard/SKILL.md
new file mode 100644
index 00000000..ac4ac6fe
--- /dev/null
+++ b/.claude/skills/openspec-onboard/SKILL.md
@@ -0,0 +1,555 @@
+---
+name: openspec-onboard
+description: Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx:propose` | Create a change and generate all artifacts |
+ | `/opsx:explore` | Think through problems before/during work  |
+ | `/opsx:apply`   | Implement tasks from a change              |
+ | `/opsx:archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx:new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx:continue` | Continue working on an existing change                   |
+ | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx:verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx:continue <name>` - Resume artifact creation
+- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx:propose <name>` | Create a change and generate all artifacts |
+ | `/opsx:explore`        | Think through problems (no code changes)   |
+ | `/opsx:apply <name>`   | Implement tasks                            |
+ | `/opsx:archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx:new <name>`      | Start a new change, step by step    |
+ | `/opsx:continue <name>` | Continue an existing change         |
+ | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx:verify <name>`   | Verify implementation               |
+
+Try `/opsx:propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.claude/skills/openspec-propose/SKILL.md b/.claude/skills/openspec-propose/SKILL.md
new file mode 100644
index 00000000..716d2d3d
--- /dev/null
+++ b/.claude/skills/openspec-propose/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: openspec-propose
+description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
+
+**Steps**
+
+1. **If no clear input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.claude/skills/openspec-sync-specs/SKILL.md b/.claude/skills/openspec-sync-specs/SKILL.md
new file mode 100644
index 00000000..b6aad67a
--- /dev/null
+++ b/.claude/skills/openspec-sync-specs/SKILL.md
@@ -0,0 +1,148 @@
+---
+name: openspec-sync-specs
+description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.claude/skills/openspec-update-change/SKILL.md b/.claude/skills/openspec-update-change/SKILL.md
new file mode 100644
index 00000000..3e1c931d
--- /dev/null
+++ b/.claude/skills/openspec-update-change/SKILL.md
@@ -0,0 +1,86 @@
+---
+name: openspec-update-change
+description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.claude/skills/openspec-verify-change/SKILL.md b/.claude/skills/openspec-verify-change/SKILL.md
new file mode 100644
index 00000000..b6557c6f
--- /dev/null
+++ b/.claude/skills/openspec-verify-change/SKILL.md
@@ -0,0 +1,172 @@
+---
+name: openspec-verify-change
+description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/.github/prompts/opsx-apply.prompt.md b/.github/prompts/opsx-apply.prompt.md
new file mode 100644
index 00000000..5bd0b6bb
--- /dev/null
+++ b/.github/prompts/opsx-apply.prompt.md
@@ -0,0 +1,152 @@
+---
+description: Implement tasks from an OpenSpec change (Experimental)
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! You can archive this change with `/opsx:archive`.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.github/prompts/opsx-archive.prompt.md b/.github/prompts/opsx-archive.prompt.md
new file mode 100644
index 00000000..f9ba89d8
--- /dev/null
+++ b/.github/prompts/opsx-archive.prompt.md
@@ -0,0 +1,157 @@
+---
+description: Archive a completed change in the experimental workflow
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Spec sync status (synced / sync skipped / no delta specs)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success (No Delta Specs)**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** No delta specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success With Warnings**
+
+```
+## Archive Complete (with warnings)
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** Sync skipped (user chose to skip)
+
+**Warnings:**
+- Archived with 2 incomplete artifacts
+- Archived with 3 incomplete tasks
+- Delta spec sync was skipped (user chose to skip)
+
+Review the archive if this was not intentional.
+```
+
+**Output On Error (Archive Exists)**
+
+```
+## Archive Failed
+
+**Change:** <change-name>
+**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+
+Target archive directory already exists.
+
+**Options:**
+1. Rename the existing archive
+2. Delete the existing archive if it's a duplicate
+3. Wait until a different date to archive
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.github/prompts/opsx-explore.prompt.md b/.github/prompts/opsx-explore.prompt.md
new file mode 100644
index 00000000..856001a6
--- /dev/null
+++ b/.github/prompts/opsx-explore.prompt.md
@@ -0,0 +1,171 @@
+---
+description: Enter explore mode - think through ideas, investigate problems, clarify requirements
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A change name: "add-dark-mode" (to explore in context of that change)
+- A comparison: "postgres vs sqlite for this"
+- Nothing (just enter explore mode)
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+If the user mentioned a specific change name, read its artifacts for context.
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.github/prompts/opsx-onboard.prompt.md b/.github/prompts/opsx-onboard.prompt.md
new file mode 100644
index 00000000..84aa48d6
--- /dev/null
+++ b/.github/prompts/opsx-onboard.prompt.md
@@ -0,0 +1,547 @@
+---
+description: Guided onboarding - walk through a complete OpenSpec workflow cycle with narration
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx:propose` | Create a change and generate all artifacts |
+ | `/opsx:explore` | Think through problems before/during work  |
+ | `/opsx:apply`   | Implement tasks from a change              |
+ | `/opsx:archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx:new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx:continue` | Continue working on an existing change                   |
+ | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx:verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx:continue <name>` - Resume artifact creation
+- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx:propose <name>` | Create a change and generate all artifacts |
+ | `/opsx:explore`        | Think through problems (no code changes)   |
+ | `/opsx:apply <name>`   | Implement tasks                            |
+ | `/opsx:archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx:new <name>`      | Start a new change, step by step    |
+ | `/opsx:continue <name>` | Continue an existing change         |
+ | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx:verify <name>`   | Verify implementation               |
+
+Try `/opsx:propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.github/prompts/opsx-propose.prompt.md b/.github/prompts/opsx-propose.prompt.md
new file mode 100644
index 00000000..fde349a6
--- /dev/null
+++ b/.github/prompts/opsx-propose.prompt.md
@@ -0,0 +1,106 @@
+---
+description: Propose a new change - create it and generate all artifacts in one step
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
+
+**Steps**
+
+1. **If no input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` to start implementing."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.github/prompts/opsx-sync.prompt.md b/.github/prompts/opsx-sync.prompt.md
new file mode 100644
index 00000000..961aa409
--- /dev/null
+++ b/.github/prompts/opsx-sync.prompt.md
@@ -0,0 +1,140 @@
+---
+description: Sync delta specs from a change to main specs
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.github/prompts/opsx-update.prompt.md b/.github/prompts/opsx-update.prompt.md
new file mode 100644
index 00000000..5b316d4d
--- /dev/null
+++ b/.github/prompts/opsx-update.prompt.md
@@ -0,0 +1,78 @@
+---
+description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.github/prompts/opsx-verify.prompt.md b/.github/prompts/opsx-verify.prompt.md
new file mode 100644
index 00000000..cc1377a7
--- /dev/null
+++ b/.github/prompts/opsx-verify.prompt.md
@@ -0,0 +1,164 @@
+---
+description: Verify implementation matches change artifacts before archiving
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/.github/skills/openspec-apply-change/SKILL.md b/.github/skills/openspec-apply-change/SKILL.md
new file mode 100644
index 00000000..1375861c
--- /dev/null
+++ b/.github/skills/openspec-apply-change/SKILL.md
@@ -0,0 +1,160 @@
+---
+name: openspec-apply-change
+description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! Ready to archive this change.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.github/skills/openspec-archive-change/SKILL.md b/.github/skills/openspec-archive-change/SKILL.md
new file mode 100644
index 00000000..c0c169d1
--- /dev/null
+++ b/.github/skills/openspec-archive-change/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: openspec-archive-change
+description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Whether specs were synced (if applicable)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
+
+All artifacts complete. All tasks complete.
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use openspec-sync-specs approach (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.github/skills/openspec-explore/SKILL.md b/.github/skills/openspec-explore/SKILL.md
new file mode 100644
index 00000000..771271a8
--- /dev/null
+++ b/.github/skills/openspec-explore/SKILL.md
@@ -0,0 +1,290 @@
+---
+name: openspec-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+      ════════════════════════════════════════════
+
+      Awareness          Coordination         Sync
+          │                   │                 │
+          ▼                   ▼                 ▼
+      ┌────────┐         ┌────────┐        ┌────────┐
+      │Presence│         │Cursors │        │  CRDT  │
+      │  "3    │         │ Multi  │        │Conflict│
+      │online" │         │ select │        │  free  │
+      └────────┘         └────────┘        └────────┘
+          │                   │                 │
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     ┌─────────────────────────────────────────────┐
+     │              CURRENT AUTH FLOW              │
+     └─────────────────────────────────────────────┘
+                        │
+          ┌─────────────┼─────────────┐
+          ▼             ▼             ▼
+     ┌─────────┐  ┌─────────┐  ┌─────────┐
+     │ Google  │  │  GitHub │  │  Email  │
+     │  OAuth  │  │  OAuth  │  │  Magic  │
+     └────┬────┘  └────┬────┘  └────┬────┘
+          │             │             │
+          └─────────────┼─────────────┘
+                        ▼
+                  ┌───────────┐
+                  │  Session  │
+                  └─────┬─────┘
+                        │
+                        ▼
+                  ┌───────────┐
+                  │   Perms   │
+                  └───────────┘
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /opsx:explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads change artifacts]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the design to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     ┌─────────────────────────────────────────────────┐
+     │          CLI TOOL DATA STORAGE                  │
+     └─────────────────────────────────────────────────┘
+
+     Key constraints:
+     • No daemon running
+     • Must work offline
+     • Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ✓      needs server ✗
+     Offline      yes ✓           no ✗
+     Single file  yes ✓           no ✗
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a change proposal
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.github/skills/openspec-onboard/SKILL.md b/.github/skills/openspec-onboard/SKILL.md
new file mode 100644
index 00000000..ac4ac6fe
--- /dev/null
+++ b/.github/skills/openspec-onboard/SKILL.md
@@ -0,0 +1,555 @@
+---
+name: openspec-onboard
+description: Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx:propose` | Create a change and generate all artifacts |
+ | `/opsx:explore` | Think through problems before/during work  |
+ | `/opsx:apply`   | Implement tasks from a change              |
+ | `/opsx:archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx:new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx:continue` | Continue working on an existing change                   |
+ | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx:verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx:continue <name>` - Resume artifact creation
+- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx:propose <name>` | Create a change and generate all artifacts |
+ | `/opsx:explore`        | Think through problems (no code changes)   |
+ | `/opsx:apply <name>`   | Implement tasks                            |
+ | `/opsx:archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx:new <name>`      | Start a new change, step by step    |
+ | `/opsx:continue <name>` | Continue an existing change         |
+ | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx:verify <name>`   | Verify implementation               |
+
+Try `/opsx:propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.github/skills/openspec-propose/SKILL.md b/.github/skills/openspec-propose/SKILL.md
new file mode 100644
index 00000000..716d2d3d
--- /dev/null
+++ b/.github/skills/openspec-propose/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: openspec-propose
+description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
+
+**Steps**
+
+1. **If no clear input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.github/skills/openspec-sync-specs/SKILL.md b/.github/skills/openspec-sync-specs/SKILL.md
new file mode 100644
index 00000000..b6aad67a
--- /dev/null
+++ b/.github/skills/openspec-sync-specs/SKILL.md
@@ -0,0 +1,148 @@
+---
+name: openspec-sync-specs
+description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.github/skills/openspec-update-change/SKILL.md b/.github/skills/openspec-update-change/SKILL.md
new file mode 100644
index 00000000..3e1c931d
--- /dev/null
+++ b/.github/skills/openspec-update-change/SKILL.md
@@ -0,0 +1,86 @@
+---
+name: openspec-update-change
+description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.github/skills/openspec-verify-change/SKILL.md b/.github/skills/openspec-verify-change/SKILL.md
new file mode 100644
index 00000000..b6557c6f
--- /dev/null
+++ b/.github/skills/openspec-verify-change/SKILL.md
@@ -0,0 +1,172 @@
+---
+name: openspec-verify-change
+description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/.gitignore b/.gitignore
index 063e5b71..fbb2f010 100644
--- a/.gitignore
+++ b/.gitignore
@@ -29,5 +29,5 @@ jest-report.json
 
 # ENV
 **/.env
-/.claude/
 /--help/
+.claude/settings.local.json
diff --git a/.opencode/commands/opsx-apply.md b/.opencode/commands/opsx-apply.md
new file mode 100644
index 00000000..f5c574ff
--- /dev/null
+++ b/.opencode/commands/opsx-apply.md
@@ -0,0 +1,152 @@
+---
+description: Implement tasks from an OpenSpec change (Experimental)
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name (e.g., `/opsx-apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx-apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx-continue`
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! You can archive this change with `/opsx-archive`.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.opencode/commands/opsx-archive.md b/.opencode/commands/opsx-archive.md
new file mode 100644
index 00000000..7f3a696f
--- /dev/null
+++ b/.opencode/commands/opsx-archive.md
@@ -0,0 +1,157 @@
+---
+description: Archive a completed change in the experimental workflow
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx-archive` (e.g., `/opsx-archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Spec sync status (synced / sync skipped / no delta specs)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success (No Delta Specs)**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** No delta specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success With Warnings**
+
+```
+## Archive Complete (with warnings)
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** Sync skipped (user chose to skip)
+
+**Warnings:**
+- Archived with 2 incomplete artifacts
+- Archived with 3 incomplete tasks
+- Delta spec sync was skipped (user chose to skip)
+
+Review the archive if this was not intentional.
+```
+
+**Output On Error (Archive Exists)**
+
+```
+## Archive Failed
+
+**Change:** <change-name>
+**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+
+Target archive directory already exists.
+
+**Options:**
+1. Rename the existing archive
+2. Delete the existing archive if it's a duplicate
+3. Wait until a different date to archive
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.opencode/commands/opsx-explore.md b/.opencode/commands/opsx-explore.md
new file mode 100644
index 00000000..928233b4
--- /dev/null
+++ b/.opencode/commands/opsx-explore.md
@@ -0,0 +1,171 @@
+---
+description: Enter explore mode - think through ideas, investigate problems, clarify requirements
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx-explore` is whatever the user wants to think about. Could be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A change name: "add-dark-mode" (to explore in context of that change)
+- A comparison: "postgres vs sqlite for this"
+- Nothing (just enter explore mode)
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+If the user mentioned a specific change name, read its artifacts for context.
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.opencode/commands/opsx-onboard.md b/.opencode/commands/opsx-onboard.md
new file mode 100644
index 00000000..3ba27d12
--- /dev/null
+++ b/.opencode/commands/opsx-onboard.md
@@ -0,0 +1,547 @@
+---
+description: Guided onboarding - walk through a complete OpenSpec workflow cycle with narration
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx-onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx-explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx-propose` | Create a change and generate all artifacts |
+ | `/opsx-explore` | Think through problems before/during work  |
+ | `/opsx-apply`   | Implement tasks from a change              |
+ | `/opsx-archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx-new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx-continue` | Continue working on an existing change                   |
+ | `/opsx-ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx-verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx-propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx-continue <name>` - Resume artifact creation
+- `/opsx-apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx-propose <name>` | Create a change and generate all artifacts |
+ | `/opsx-explore`        | Think through problems (no code changes)   |
+ | `/opsx-apply <name>`   | Implement tasks                            |
+ | `/opsx-archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx-new <name>`      | Start a new change, step by step    |
+ | `/opsx-continue <name>` | Continue an existing change         |
+ | `/opsx-ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx-verify <name>`   | Verify implementation               |
+
+Try `/opsx-propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.opencode/commands/opsx-propose.md b/.opencode/commands/opsx-propose.md
new file mode 100644
index 00000000..6b498c26
--- /dev/null
+++ b/.opencode/commands/opsx-propose.md
@@ -0,0 +1,106 @@
+---
+description: Propose a new change - create it and generate all artifacts in one step
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx-apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx-propose` is the change name (kebab-case), OR a description of what the user wants to build.
+
+**Steps**
+
+1. **If no input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx-apply` to start implementing."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.opencode/commands/opsx-sync.md b/.opencode/commands/opsx-sync.md
new file mode 100644
index 00000000..5f1b7bb4
--- /dev/null
+++ b/.opencode/commands/opsx-sync.md
@@ -0,0 +1,140 @@
+---
+description: Sync delta specs from a change to main specs
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx-sync` (e.g., `/opsx-sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.opencode/commands/opsx-update.md b/.opencode/commands/opsx-update.md
new file mode 100644
index 00000000..7327a92f
--- /dev/null
+++ b/.opencode/commands/opsx-update.md
@@ -0,0 +1,78 @@
+---
+description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx-update` (e.g., `/opsx-update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx-continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx-continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx-apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx-archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx-continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx-apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx-continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx-new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.opencode/commands/opsx-verify.md b/.opencode/commands/opsx-verify.md
new file mode 100644
index 00000000..de6d1450
--- /dev/null
+++ b/.opencode/commands/opsx-verify.md
@@ -0,0 +1,164 @@
+---
+description: Verify implementation matches change artifacts before archiving
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx-verify` (e.g., `/opsx-verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/.opencode/skills/openspec-apply-change/SKILL.md b/.opencode/skills/openspec-apply-change/SKILL.md
new file mode 100644
index 00000000..e44e01f9
--- /dev/null
+++ b/.opencode/skills/openspec-apply-change/SKILL.md
@@ -0,0 +1,160 @@
+---
+name: openspec-apply-change
+description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx-apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! Ready to archive this change.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.opencode/skills/openspec-archive-change/SKILL.md b/.opencode/skills/openspec-archive-change/SKILL.md
new file mode 100644
index 00000000..c0c169d1
--- /dev/null
+++ b/.opencode/skills/openspec-archive-change/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: openspec-archive-change
+description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Whether specs were synced (if applicable)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
+
+All artifacts complete. All tasks complete.
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use openspec-sync-specs approach (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.opencode/skills/openspec-explore/SKILL.md b/.opencode/skills/openspec-explore/SKILL.md
new file mode 100644
index 00000000..894ef32e
--- /dev/null
+++ b/.opencode/skills/openspec-explore/SKILL.md
@@ -0,0 +1,290 @@
+---
+name: openspec-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+      ════════════════════════════════════════════
+
+      Awareness          Coordination         Sync
+          │                   │                 │
+          ▼                   ▼                 ▼
+      ┌────────┐         ┌────────┐        ┌────────┐
+      │Presence│         │Cursors │        │  CRDT  │
+      │  "3    │         │ Multi  │        │Conflict│
+      │online" │         │ select │        │  free  │
+      └────────┘         └────────┘        └────────┘
+          │                   │                 │
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     ┌─────────────────────────────────────────────┐
+     │              CURRENT AUTH FLOW              │
+     └─────────────────────────────────────────────┘
+                        │
+          ┌─────────────┼─────────────┐
+          ▼             ▼             ▼
+     ┌─────────┐  ┌─────────┐  ┌─────────┐
+     │ Google  │  │  GitHub │  │  Email  │
+     │  OAuth  │  │  OAuth  │  │  Magic  │
+     └────┬────┘  └────┬────┘  └────┬────┘
+          │             │             │
+          └─────────────┼─────────────┘
+                        ▼
+                  ┌───────────┐
+                  │  Session  │
+                  └─────┬─────┘
+                        │
+                        ▼
+                  ┌───────────┐
+                  │   Perms   │
+                  └───────────┘
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /opsx-explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads change artifacts]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the design to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     ┌─────────────────────────────────────────────────┐
+     │          CLI TOOL DATA STORAGE                  │
+     └─────────────────────────────────────────────────┘
+
+     Key constraints:
+     • No daemon running
+     • Must work offline
+     • Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ✓      needs server ✗
+     Offline      yes ✓           no ✗
+     Single file  yes ✓           no ✗
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a change proposal
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/.opencode/skills/openspec-onboard/SKILL.md b/.opencode/skills/openspec-onboard/SKILL.md
new file mode 100644
index 00000000..52c4228f
--- /dev/null
+++ b/.opencode/skills/openspec-onboard/SKILL.md
@@ -0,0 +1,555 @@
+---
+name: openspec-onboard
+description: Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## Preflight
+
+Before starting, check if the OpenSpec CLI is installed:
+
+```bash
+# Unix/macOS
+openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
+# Windows (PowerShell)
+# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
+```
+
+**If CLI not installed:**
+> OpenSpec CLI is not installed. Install it first, then come back to `/opsx-onboard`.
+
+Stop here if not installed.
+
+---
+
+## Phase 1: Welcome
+
+Display:
+
+```
+## Welcome to OpenSpec!
+
+I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
+
+**What we'll do:**
+1. Pick a small, real task in your codebase
+2. Explore the problem briefly
+3. Create a change (the container for our work)
+4. Build the artifacts: proposal → specs → design → tasks
+5. Implement the tasks
+6. Archive the completed change
+
+**Time:** ~15-20 minutes
+
+Let's start by finding something to work on.
+```
+
+---
+
+## Phase 2: Task Selection
+
+### Codebase Analysis
+
+Scan the codebase for small improvement opportunities. Look for:
+
+1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
+2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
+3. **Functions without tests** - Cross-reference `src/` with test directories
+4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
+5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
+6. **Missing validation** - User input handlers without validation
+
+Also check recent git activity:
+```bash
+# Unix/macOS
+git log --oneline -10 2>/dev/null || echo "No git history"
+# Windows (PowerShell)
+# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
+```
+
+### Present Suggestions
+
+From your analysis, present 3-4 specific suggestions:
+
+```
+## Task Suggestions
+
+Based on scanning your codebase, here are some good starter tasks:
+
+**1. [Most promising task]**
+   Location: `src/path/to/file.ts:42`
+   Scope: ~1-2 files, ~20-30 lines
+   Why it's good: [brief reason]
+
+**2. [Second task]**
+   Location: `src/another/file.ts`
+   Scope: ~1 file, ~15 lines
+   Why it's good: [brief reason]
+
+**3. [Third task]**
+   Location: [location]
+   Scope: [estimate]
+   Why it's good: [brief reason]
+
+**4. Something else?**
+   Tell me what you'd like to work on.
+
+Which task interests you? (Pick a number or describe your own)
+```
+
+**If nothing found:** Fall back to asking what the user wants to build:
+> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
+
+### Scope Guardrail
+
+If the user picks or describes something too large (major feature, multi-day work):
+
+```
+That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
+
+For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
+
+**Options:**
+1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
+2. **Pick something else** - One of the other suggestions, or a different small task?
+3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
+
+What would you prefer?
+```
+
+Let the user override if they insist—this is a soft guardrail.
+
+---
+
+## Phase 3: Explore Demo
+
+Once a task is selected, briefly demonstrate explore mode:
+
+```
+Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
+```
+
+Spend 1-2 minutes investigating the relevant code:
+- Read the file(s) involved
+- Draw a quick ASCII diagram if it helps
+- Note any considerations
+
+```
+## Quick Exploration
+
+[Your brief analysis—what you found, any considerations]
+
+┌─────────────────────────────────────────┐
+│   [Optional: ASCII diagram if helpful]  │
+└─────────────────────────────────────────┘
+
+Explore mode (`/opsx-explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
+
+Now let's create a change to hold our work.
+```
+
+**PAUSE** - Wait for user acknowledgment before proceeding.
+
+---
+
+## Phase 4: Create the Change
+
+**EXPLAIN:**
+```
+## Creating a Change
+
+A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
+
+Let me create one for our task.
+```
+
+**DO:** Create the change with a derived kebab-case name:
+```bash
+openspec new change "<derived-name>"
+```
+
+**SHOW:**
+```
+Created: <changeRoot from status JSON>
+
+The folder structure:
+```
+<changeRoot>/
+├── proposal.md    ← Why we're doing this (empty, we'll fill it)
+├── design.md      ← How we'll build it (empty)
+├── specs/         ← Detailed requirements (empty)
+└── tasks.md       ← Implementation checklist (empty)
+```
+
+Now let's fill in the first artifact—the proposal.
+```
+
+---
+
+## Phase 5: Proposal
+
+**EXPLAIN:**
+```
+## The Proposal
+
+The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
+
+I'll draft one based on our task.
+```
+
+**DO:** Draft the proposal content (don't save yet):
+
+```
+Here's a draft proposal:
+
+---
+
+## Why
+
+[1-2 sentences explaining the problem/opportunity]
+
+## What Changes
+
+[Bullet points of what will be different]
+
+## Capabilities
+
+### New Capabilities
+- `<capability-name>`: [brief description]
+
+### Modified Capabilities
+<!-- If modifying existing behavior -->
+
+## Impact
+
+- `src/path/to/file.ts`: [what changes]
+- [other files if applicable]
+
+---
+
+Does this capture the intent? I can adjust before we save it.
+```
+
+**PAUSE** - Wait for user approval/feedback.
+
+After approval, save the proposal:
+```bash
+openspec instructions proposal --change "<name>" --json
+```
+Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
+
+```
+Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
+
+Next up: specs.
+```
+
+---
+
+## Phase 6: Specs
+
+**EXPLAIN:**
+```
+## Specs
+
+Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
+
+For a small task like this, we might only need one spec file.
+```
+
+**DO:** Resolve where the spec file should be created:
+```bash
+openspec instructions specs --change "<name>" --json
+# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
+```
+
+Draft the spec content:
+
+```
+Here's the spec:
+
+---
+
+## ADDED Requirements
+
+### Requirement: <Name>
+
+<Description of what the system should do>
+
+#### Scenario: <Scenario name>
+
+- **WHEN** <trigger condition>
+- **THEN** <expected outcome>
+- **AND** <additional outcome if needed>
+
+---
+
+This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
+```
+
+Save to the concrete file path chosen from `resolvedOutputPath`.
+
+---
+
+## Phase 7: Design
+
+**EXPLAIN:**
+```
+## Design
+
+The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
+
+For small changes, this might be brief. That's fine—not every change needs deep design discussion.
+```
+
+**DO:** Draft design.md:
+
+```
+Here's the design:
+
+---
+
+## Context
+
+[Brief context about the current state]
+
+## Goals / Non-Goals
+
+**Goals:**
+- [What we're trying to achieve]
+
+**Non-Goals:**
+- [What's explicitly out of scope]
+
+## Decisions
+
+### Decision 1: [Key decision]
+
+[Explanation of approach and rationale]
+
+---
+
+For a small task, this captures the key decisions without over-engineering.
+```
+
+Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
+
+---
+
+## Phase 8: Tasks
+
+**EXPLAIN:**
+```
+## Tasks
+
+Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
+
+These should be small, clear, and in logical order.
+```
+
+**DO:** Generate tasks based on specs and design:
+
+```
+Here are the implementation tasks:
+
+---
+
+## 1. [Category or file]
+
+- [ ] 1.1 [Specific task]
+- [ ] 1.2 [Specific task]
+
+## 2. Verify
+
+- [ ] 2.1 [Verification step]
+
+---
+
+Each checkbox becomes a unit of work in the apply phase. Ready to implement?
+```
+
+**PAUSE** - Wait for user to confirm they're ready to implement.
+
+Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
+
+---
+
+## Phase 9: Apply (Implementation)
+
+**EXPLAIN:**
+```
+## Implementation
+
+Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
+```
+
+**DO:** For each task:
+
+1. Announce: "Working on task N: [description]"
+2. Implement the change in the codebase
+3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
+4. Mark complete in tasks.md: `- [ ]` → `- [x]`
+5. Brief status: "✓ Task N complete"
+
+Keep narration light—don't over-explain every line of code.
+
+After all tasks:
+
+```
+## Implementation Complete
+
+All tasks done:
+- [x] Task 1
+- [x] Task 2
+- [x] ...
+
+The change is implemented! One more step—let's archive it.
+```
+
+---
+
+## Phase 10: Archive
+
+**EXPLAIN:**
+```
+## Archiving
+
+When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
+
+Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
+```
+
+**DO:**
+```bash
+openspec archive "<name>"
+```
+
+**SHOW:**
+```
+Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
+
+The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
+```
+
+---
+
+## Phase 11: Recap & Next Steps
+
+```
+## Congratulations!
+
+You just completed a full OpenSpec cycle:
+
+1. **Explore** - Thought through the problem
+2. **New** - Created a change container
+3. **Proposal** - Captured WHY
+4. **Specs** - Defined WHAT in detail
+5. **Design** - Decided HOW
+6. **Tasks** - Broke it into steps
+7. **Apply** - Implemented the work
+8. **Archive** - Preserved the record
+
+This same rhythm works for any size change—a small fix or a major feature.
+
+---
+
+## Command Reference
+
+**Core workflow:**
+
+ | Command           | What it does                               |
+ |-------------------|--------------------------------------------|
+ | `/opsx-propose` | Create a change and generate all artifacts |
+ | `/opsx-explore` | Think through problems before/during work  |
+ | `/opsx-apply`   | Implement tasks from a change              |
+ | `/opsx-archive` | Archive a completed change                 |
+
+**Additional commands:**
+
+ | Command            | What it does                                             |
+ |--------------------|----------------------------------------------------------|
+ | `/opsx-new`      | Start a new change, step through artifacts one at a time |
+ | `/opsx-continue` | Continue working on an existing change                   |
+ | `/opsx-ff`       | Fast-forward: create all artifacts at once               |
+ | `/opsx-verify`   | Verify implementation matches artifacts                  |
+
+---
+
+## What's Next?
+
+Try `/opsx-propose` on something you actually want to build. You've got the rhythm now!
+```
+
+---
+
+## Graceful Exit Handling
+
+### User wants to stop mid-way
+
+If the user says they need to stop, want to pause, or seem disengaged:
+
+```
+No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
+
+To pick up where we left off later:
+- `/opsx-continue <name>` - Resume artifact creation
+- `/opsx-apply <name>` - Jump to implementation (if tasks exist)
+
+The work won't be lost. Come back whenever you're ready.
+```
+
+Exit gracefully without pressure.
+
+### User just wants command reference
+
+If the user says they just want to see the commands or skip the tutorial:
+
+```
+## OpenSpec Quick Reference
+
+**Core workflow:**
+
+ | Command                  | What it does                               |
+ |--------------------------|--------------------------------------------|
+ | `/opsx-propose <name>` | Create a change and generate all artifacts |
+ | `/opsx-explore`        | Think through problems (no code changes)   |
+ | `/opsx-apply <name>`   | Implement tasks                            |
+ | `/opsx-archive <name>` | Archive when done                          |
+
+**Additional commands:**
+
+ | Command                   | What it does                        |
+ |---------------------------|-------------------------------------|
+ | `/opsx-new <name>`      | Start a new change, step by step    |
+ | `/opsx-continue <name>` | Continue an existing change         |
+ | `/opsx-ff <name>`       | Fast-forward: all artifacts at once |
+ | `/opsx-verify <name>`   | Verify implementation               |
+
+Try `/opsx-propose` to start your first change.
+```
+
+Exit gracefully.
+
+---
+
+## Guardrails
+
+- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
+- **Keep narration light** during implementation—teach without lecturing
+- **Don't skip phases** even if the change is small—the goal is teaching the workflow
+- **Pause for acknowledgment** at marked points, but don't over-pause
+- **Handle exits gracefully**—never pressure the user to continue
+- **Use real codebase tasks**—don't simulate or use fake examples
+- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.opencode/skills/openspec-propose/SKILL.md b/.opencode/skills/openspec-propose/SKILL.md
new file mode 100644
index 00000000..f0af3e15
--- /dev/null
+++ b/.opencode/skills/openspec-propose/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: openspec-propose
+description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx-apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
+
+**Steps**
+
+1. **If no clear input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx-apply` or ask me to implement to start working on the tasks."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/.opencode/skills/openspec-sync-specs/SKILL.md b/.opencode/skills/openspec-sync-specs/SKILL.md
new file mode 100644
index 00000000..b6aad67a
--- /dev/null
+++ b/.opencode/skills/openspec-sync-specs/SKILL.md
@@ -0,0 +1,148 @@
+---
+name: openspec-sync-specs
+description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/.opencode/skills/openspec-update-change/SKILL.md b/.opencode/skills/openspec-update-change/SKILL.md
new file mode 100644
index 00000000..efa37f4c
--- /dev/null
+++ b/.opencode/skills/openspec-update-change/SKILL.md
@@ -0,0 +1,86 @@
+---
+name: openspec-update-change
+description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx-continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx-continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx-apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx-archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx-continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx-apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx-continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx-new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.opencode/skills/openspec-verify-change/SKILL.md b/.opencode/skills/openspec-verify-change/SKILL.md
new file mode 100644
index 00000000..b6557c6f
--- /dev/null
+++ b/.opencode/skills/openspec-verify-change/SKILL.md
@@ -0,0 +1,172 @@
+---
+name: openspec-verify-change
+description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/openspec/config.yaml b/openspec/config.yaml
new file mode 100644
index 00000000..392946c6
--- /dev/null
+++ b/openspec/config.yaml
@@ -0,0 +1,20 @@
+schema: spec-driven
+
+# Project context (optional)
+# This is shown to AI when creating artifacts.
+# Add your tech stack, conventions, style guides, domain knowledge, etc.
+# Example:
+#   context: |
+#     Tech stack: TypeScript, React, Node.js
+#     We use conventional commits
+#     Domain: e-commerce platform
+
+# Per-artifact rules (optional)
+# Add custom rules for specific artifacts.
+# Example:
+#   rules:
+#     proposal:
+#       - Keep proposals under 500 words
+#       - Always include a "Non-goals" section
+#     tasks:
+#       - Break tasks into chunks of max 2 hours
diff --git a/openspec/specs/architecture/spec.md b/openspec/specs/architecture/spec.md
new file mode 100644
index 00000000..28c0161f
--- /dev/null
+++ b/openspec/specs/architecture/spec.md
@@ -0,0 +1,77 @@
+# Architecture
+
+## Purpose
+
+Cross-cutting structural invariants that span multiple packages and aren't owned by any single capability spec: the dependency direction between packages, the event-channel naming convention used across the `EventBus`, the single-adapter contract `core` enforces, and which undo/redo implementation takes precedence when collaboration is enabled. These constrain how packages may depend on and communicate with each other; package-specific behavior lives in the respective capability specs ([[model-types]], [[sdk]], [[model]], [[dom-adapters]], [[collaboration-manager]], [[core]], [[ui]], [[ot-server]], [[tools]]).
+
+## Requirements
+
+### Requirement: Tools and plugins depend on the SDK, not on model or core
+Tool and plugin packages SHALL depend on `@editorjs/sdk` (and, where DOM binding is needed, `@editorjs/dom-adapters`) for their contracts, and SHALL NOT declare a direct dependency on `@editorjs/model` or `@editorjs/core`.
+
+#### Scenario: Built-in tool package dependencies
+- **GIVEN** the `package.json` of a built-in tool (`@editorjs/paragraph`, `@editorjs/bold`, `@editorjs/italic`, `@editorjs/inline-link`)
+- **WHEN** its `dependencies` are inspected
+- **THEN** it lists `@editorjs/sdk` (and optionally `@editorjs/dom-adapters`) but never `@editorjs/model` or `@editorjs/core`
+
+#### Scenario: model-types is internal to model and sdk
+- **GIVEN** `@editorjs/model-types` is a dependency-free, internal-only package
+- **WHEN** any package other than `@editorjs/model` or `@editorjs/sdk` needs its types
+- **THEN** it SHALL obtain them indirectly through `@editorjs/sdk`'s re-exports rather than depending on `@editorjs/model-types` directly
+
+Confirmed via `packages/tools/{paragraph,bold,italic,inline-link}/package.json` and `packages/sdk/package.json`.
+
+### Requirement: EventBus channel naming convention
+Every event dispatched on an `EventBus` SHALL be namespaced by its origin as a `<channel>:<name>` string, using one of the fixed channel prefixes `core:` (orchestrator-level events), `ui:` (UI-shell events), or `adapter:` (DOM-adapter events). Event classes SHALL apply the prefix in their base class rather than each concrete event repeating it.
+
+#### Scenario: Core events are auto-prefixed
+- **GIVEN** a concrete event class extends `CoreEventBase` with a bare name such as `block:added`
+- **WHEN** the event is constructed
+- **THEN** it dispatches as `core:block:added`, since `CoreEventBase`'s constructor prepends `core:` to the given name
+
+#### Scenario: UI events are auto-prefixed
+- **GIVEN** a concrete event class extends `UIEventBase` with a bare name such as `toolbar:rendered`
+- **WHEN** the event is constructed
+- **THEN** it dispatches as `ui:toolbar:rendered`, since `UIEventBase`'s constructor prepends `ui:` to the given name
+
+#### Scenario: Adapter events use a flat channel
+- **GIVEN** the `AdapterEventType.Updated` enum value
+- **WHEN** an adapter event is dispatched
+- **THEN** its type is the literal `adapter:updated`, since adapter events are not built on a shared prefixing base class
+
+Implemented in `packages/sdk/src/entities/EventBus/events/core/CoreEventBase.ts`, `.../ui/UIEventBase.ts`, `.../core/CoreEventType.ts`, `.../adapter/AdapterEventType.ts`.
+
+### Requirement: Exactly one adapter plugin is bound
+`Core` SHALL require exactly one `PluginType.Adapter` plugin bound in its plugin container at initialization time. Registering a second adapter via `use()` SHALL replace the previously bound one rather than error or bind both.
+
+#### Scenario: Registering a second adapter replaces the first
+- **GIVEN** an adapter plugin has already been registered via `core.use(SomeAdapter)`
+- **WHEN** `core.use(AnotherAdapter)` is called with another `PluginType.Adapter` plugin
+- **THEN** the container rebinds `PluginType.Adapter` to the new adapter, so only the most recently registered adapter is active when `#initializeAdapter()` runs
+
+#### Scenario: Default adapter is DOMAdapters
+- **GIVEN** a `Core` instance is constructed without an explicit adapter registration
+- **WHEN** the constructor runs
+- **THEN** it calls `this.use(DOMAdapters)`, so `@editorjs/dom-adapters` is bound as the default adapter
+
+Implemented in `packages/core/src/index.ts` (`use()`, `#initializeAdapter()`), `packages/core/src/tokens.ts` (`TOKENS.Adapter`).
+
+### Requirement: Collaboration's undo/redo preempts core's local undo/redo
+When `@editorjs/collaboration-manager` is registered, it SHALL intercept `core:undo`/`core:redo` events and call `preventDefault()` on them before `core`'s own `UndoRedoManager` acts, substituting its own OT-aware undo/redo (which accounts for remote operations) for `core`'s local-only undo/redo. This is an intentional override, not a duplicated/competing implementation.
+
+#### Scenario: Collaboration manager suppresses core's default undo
+- **GIVEN** both `core`'s `UndoRedoManager` and `CollaborationManager` are listening for `core:undo`
+- **WHEN** a `core:undo` event is dispatched (e.g. via Cmd/Ctrl+Z)
+- **THEN** `CollaborationManager`'s listener calls `e.preventDefault()` and performs its own OT-aware `undo()`, so `core`'s local `UndoRedoManager` does not also apply an undo for the same event
+
+#### Scenario: Redo follows the same override
+- **GIVEN** the same dual-listener setup
+- **WHEN** a `core:redo` event is dispatched
+- **THEN** `CollaborationManager` calls `e.preventDefault()` and performs its own `redo()` instead of `core`'s default
+
+#### Scenario: Without collaboration-manager, core's undo/redo is authoritative
+- **GIVEN** `@editorjs/collaboration-manager` is not registered
+- **WHEN** a `core:undo`/`core:redo` event is dispatched
+- **THEN** no listener calls `preventDefault()`, so `core`'s own `UndoRedoManager` performs the undo/redo
+
+Implemented in `packages/collaboration-manager/src/CollaborationManager.ts` (constructor's `onUndo`/`onRedo` handlers), `packages/core/src/components/UndoRedoManager.ts`.
diff --git a/openspec/specs/collaboration-manager/spec.md b/openspec/specs/collaboration-manager/spec.md
new file mode 100644
index 00000000..ef7e19ac
--- /dev/null
+++ b/openspec/specs/collaboration-manager/spec.md
@@ -0,0 +1,107 @@
+# Collaboration Manager
+
+## Purpose
+
+`@editorjs/collaboration-manager` is a core plugin that listens to document model events, converts them into operational-transformation (OT) `Operation`s, batches rapid local edits for undo/redo, maintains local undo/redo stacks transformed against remote edits, and optionally syncs with a remote OT server over WebSocket using a handshake + serialized-ack protocol.
+
+## Requirements
+
+### Requirement: Operation model
+The system SHALL represent every document mutation as an `Operation` (Insert/Delete/Modify/Neutral) supporting `inverse()`, `transform()` against another operation, `serialize()`, and `getEffectiveRange()`.
+
+#### Scenario: Inverting an operation
+- **GIVEN** an `Insert` or `Delete` `Operation`
+- **WHEN** `inverse()` is called
+- **THEN** it returns the complementary operation (e.g. a `Delete` for an `Insert`) that undoes the original's effect
+
+Implemented in `src/Operation.ts`, validated by `src/Operation.spec.ts`.
+
+### Requirement: Operation transformation rules
+The system SHALL implement OT transform rules for colliding block/text/data index operations (insert vs. insert, insert vs. delete, modify vs. modify, etc.), revision-aware for data operations, via `OperationsTransformer`.
+
+#### Scenario: Transforming two colliding operations
+- **GIVEN** two operations that touch overlapping or adjacent indexes
+- **WHEN** `OperationsTransformer` transforms one against the other
+- **THEN** it returns an adjusted operation whose index/range accounts for the other operation's effect, per the OT rule for that combination of operation types
+
+Implemented in `src/OperationsTransformer.ts`, validated by `src/OperationsTransformer.spec.ts`.
+
+### Requirement: Batching rapid local edits
+The system SHALL group consecutive character-level text operations from the same user into a single `BatchedOperation` for undo/redo purposes, using a debounce window before committing the batch.
+
+#### Scenario: Debounced batching
+- **GIVEN** a user types several characters in quick succession
+- **WHEN** less than 500ms elapses between keystrokes
+- **THEN** no `put` call is made to the undo/redo manager until 500ms of inactivity elapses, after which a single `BatchedOperation` covering all the characters is committed
+
+#### Scenario: Undoing a batch
+- **GIVEN** a committed `BatchedOperation` covering several consecutive inserts
+- **WHEN** `undo()` is called
+- **THEN** the entire batch is reverted at once, not character by character
+
+Implemented in `src/BatchedOperation.ts`, `src/CollaborationManager.ts`, validated by `src/BatchedOperation.spec.ts`, `src/CollaborationManager.spec.ts`.
+
+### Requirement: Remote operations transform the in-progress local batch
+The system SHALL transform an open (not-yet-committed) local batch against any remote operation that arrives while it is open, rather than ignoring the remote change until the batch closes.
+
+#### Scenario: Remote insert merges into an open local batch
+- **GIVEN** a local insert of "test" at index [0,4] is still open (not yet debounced/committed)
+- **WHEN** a remote insert of "hello" at index [0,5] arrives
+- **THEN** the local batch is transformed so the resulting text is "hellotest"
+
+#### Scenario: Conflicting remote delete clears the local batch
+- **GIVEN** an open local batch
+- **WHEN** a remote delete operation overlaps the batch's effective range such that it cannot be transformed
+- **THEN** the batch is reduced to Neutral and discarded, leaving nothing for undo to revert
+
+### Requirement: Undo/redo stack rebasing against remote edits
+The system SHALL maintain local undo/redo stacks (`UndoRedoManager`) and rebase (`transformStacks()`) them against incoming remote operations, dropping stack entries that become Neutral as a result.
+
+#### Scenario: Undo scope depends on overlap with remote edits
+- **GIVEN** a local edit "world" was typed and debounced (committed to the undo stack)
+- **WHEN** a remote insert of "hello" at index 0 arrives before the local user calls `undo()`
+- **THEN** `undo()` removes only "world", leaving "hello" — because the stacks were rebased against the remote operation before the undo was applied
+
+#### Scenario: Undo scope narrows when a remote edit lands inside a char-by-char batch
+- **GIVEN** a local batch built from individual character inserts
+- **WHEN** a remote insert lands inside that batch's range
+- **THEN** only the portion of the batch not affected by the remote insert is undone
+
+#### Scenario: Undo scope covers the whole insert when unbatched
+- **GIVEN** a single (non-batched) local insert operation
+- **WHEN** a remote insert lands inside its range
+- **THEN** the entire local insert is undone as one unit
+
+Implemented in `src/UndoRedoManager.ts`, validated by `src/UndoRedoManager.spec.ts`.
+
+### Requirement: OT WebSocket client
+The system SHALL provide `OTClient`, which performs a handshake with the OT server, serializes outgoing operations behind an acknowledgement gate, and transforms incoming remote operations against any still-pending local operations before delivering them.
+
+#### Scenario: Serialized send with ack-gating
+- **GIVEN** a local operation has been sent and is awaiting server acknowledgement
+- **WHEN** a second operation is queued via `send()`
+- **THEN** it remains queued and is not transmitted until the first operation's acknowledgement arrives, at which point the revision counter increments and the next queued operation is sent
+
+#### Scenario: Filtering and transforming incoming remote operations
+- **GIVEN** a remote operation message arrives while local operations are pending/unacknowledged
+- **WHEN** the message is processed
+- **THEN** it is transformed against the pending local operations before being surfaced via `onRemoteOperation`; if the transform reduces it to Neutral, the callback is not invoked; messages originating from the client's own `userId`, or of a non-`Operation` message type, are ignored outright
+
+#### Scenario: Cleanup
+- **GIVEN** an active `CollaborationManager` with a connected `OTClient`
+- **WHEN** `destroy()` is called
+- **THEN** the WebSocket connection is closed and all model/event-bus listeners are unsubscribed
+
+Implemented in `src/client/OTClient.ts`, `src/client/Message.ts`, `src/client/MessageType.ts`, validated by `src/client/OTClient.spec.ts`.
+
+**Known limitation**: reconnect/retry on WebSocket disconnect is not implemented (`src/client/OTClient.ts` carries `@todo` markers for offline handling and retries); no test exercises reconnect behavior.
+
+### Requirement: Range intersection classification
+The system SHALL classify how two text ranges relate (None/Left/Right/Includes/IncludedBy) to support transform and batch-overlap logic.
+
+#### Scenario: Classifying two ranges
+- **GIVEN** two text ranges
+- **WHEN** `getRangesIntersectionType` is called with them
+- **THEN** it returns one of `None`, `Left`, `Right`, `Includes`, or `IncludedBy` describing how the ranges overlap
+
+Implemented in `src/utils/getRangesIntersectionType.ts`.
diff --git a/openspec/specs/core/spec.md b/openspec/specs/core/spec.md
new file mode 100644
index 00000000..7c9d4180
--- /dev/null
+++ b/openspec/specs/core/spec.md
@@ -0,0 +1,102 @@
+# Core
+
+## Purpose
+
+`@editorjs/core` is the Editor.js orchestrator: an IoC-based composition root (`Core` class) that wires up the document model, tool/plugin lifecycle, and the `EditorAPI` surface consumed by tools, plugins, and adapters. It owns block rendering, selection/caret tracking, and local undo/redo, and requires exactly one registered adapter plugin to bind to the DOM layer.
+
+## Requirements
+
+### Requirement: Core composition root
+The system SHALL provide a `Core` class owning two IoC containers (one for singleton services, one for registered tools/plugins), exposing `use()` to register tools/plugins/adapters and `initialize()` to boot the editor.
+
+#### Scenario: Initialization order
+- **GIVEN** tools, plugins, and an adapter have been registered via `use()`
+- **WHEN** `initialize()` is called
+- **THEN** `SelectionManager`, `BlocksManager`, `BlockRenderer`, and `UndoRedoManager` are resolved from the IoC container, plugins are initialized, tools are initialized, the model's document is initialized, and finally a `CoreEventType.Ready` event is dispatched
+
+#### Scenario: Exactly one adapter is required
+- **GIVEN** one or more `PluginType.Adapter` plugins are registered via `use()`
+- **WHEN** the adapter binding is resolved
+- **THEN** the last registered adapter wins (via `rebind`) and is lazily resolved into `BlockRenderer` under `TOKENS.Adapter`
+
+Implemented in `src/index.ts`, `src/tokens.ts`.
+
+### Requirement: Tool registration and validation
+The system SHALL validate configured tools via `ToolsManager`, calling each tool's static `prepare()`, and sorting tools into `available`/`unavailable` collections exposed as `blockTools`/`inlineTools`/`blockTunes`.
+
+#### Scenario: Rejecting a malformed tool
+- **GIVEN** a configured tool is neither a constructor function nor an object with a `class` property
+- **WHEN** `ToolsManager` validates the configuration
+- **THEN** it throws an error naming the offending tool and describing the required shape
+
+Implemented in `src/tools/ToolsManager.ts`, `src/tools/ToolsFactory.ts`, validated by co-located `.spec.ts` files.
+
+### Requirement: EditorAPI surface
+The system SHALL expose an `EditorAPI` aggregating `BlocksAPI` (insert/insertMany/delete/move/render/clear/getBlocksCount), `SelectionAPI` (applyInlineTool), `DocumentAPI` (serialized data, onUpdate, insertData/removeData/modifyData, undo/redo), and `TextAPI` (insert/remove/format/unformat/getFragments/get) to tools, plugins, and adapters.
+
+#### Scenario: Deleting with no block selected
+- **GIVEN** no block is currently selected/no caret is set
+- **WHEN** `BlocksAPI.delete()` or `move()` is called without an explicit index
+- **THEN** it throws an error ("No block selected to delete" / "No block selected to move")
+
+#### Scenario: Splitting or converting with an unknown tool
+- **GIVEN** a `splitBlock`/`convertBlock` call references a `dataKey` that doesn't exist, or a source/target tool that isn't registered
+- **WHEN** the operation is attempted
+- **THEN** it throws an error identifying the missing key or tool
+
+Implemented in `src/api/index.ts`, `src/api/BlocksAPI.ts`, `src/api/SelectionAPI.ts`, `src/api/DocumentAPI/DocumentAPI.ts`, `src/api/TextAPI.ts`, `src/components/BlockManager.ts`, validated by co-located `.spec.ts`/`.integration.spec.ts` files.
+
+### Requirement: Block rendering lifecycle
+The system SHALL provide `BlockRenderer`, which listens for model `BlockAddedEvent`/`BlockRemovedEvent`, creates a `BlockToolAdapter` per block, instantiates and renders the corresponding tool, and dispatches `BlockAddedCoreEvent`/`BlockRemovedCoreEvent`.
+
+#### Scenario: Malformed or unknown block events
+- **GIVEN** a block event with an undefined index, or referencing a tool name that isn't registered
+- **WHEN** `BlockRenderer` processes the event
+- **THEN** it throws `[BlockRenderer] Block index should be defined...` or `[BlockRenderer] Block Tool <name> not found` respectively
+
+#### Scenario: Tool render failure is logged, not thrown
+- **GIVEN** a tool's `render()` method returns a rejected promise
+- **WHEN** `BlockRenderer` renders that block
+- **THEN** the rejection is logged rather than propagated as an uncaught error
+
+Implemented in `src/components/BlockRenderer.ts`, validated by its co-located `.spec.ts`.
+
+### Requirement: Selection and inline formatting
+The system SHALL provide `SelectionManager`, which tracks caret/selection state and applies inline tools to the current selection via `applyInlineTool()`.
+
+#### Scenario: Applying an inline tool with invalid state
+- **GIVEN** the caret is unset/null, the requested tool isn't found, or a selection segment lacks `textRange`/`blockIndex`/`dataKey`
+- **WHEN** `applyInlineTool()` is called
+- **THEN** it throws an error identifying the missing state
+
+#### Scenario: Formatting dispatch
+- **GIVEN** a valid selection and a registered inline tool
+- **WHEN** `applyInlineTool()` is called
+- **THEN** it calls `model.format` or `model.unformat` depending on the tool's `getFormattingOptions` result
+
+Implemented in `src/components/SelectionManager.ts`, validated by its co-located `.spec.ts`.
+
+### Requirement: Local undo/redo
+The system SHALL provide `UndoRedoManager`, which batches consecutive model events (debounced) into a single undo step, inverts `Added`/`Removed`/`Modified` events on `undo()`, and responds to cancellable `core:undo`/`core:redo` events.
+
+#### Scenario: In-progress batch flushed before undo
+- **GIVEN** an undo step is mid-batch when `undo()` is triggered
+- **WHEN** the undo is processed
+- **THEN** the in-progress batch is flushed first, and the manager ignores model events it triggers itself while replaying the undo
+
+#### Scenario: Cancellable undo/redo events
+- **GIVEN** a listener calls `preventDefault()` (or equivalent cancellation) on a `core:undo`/`core:redo` event
+- **WHEN** the event is dispatched
+- **THEN** the default undo/redo behavior is suppressed
+
+Implemented in `src/components/UndoRedoManager.ts`, validated by its co-located `.spec.ts`.
+
+### Requirement: Keyboard shortcuts plugin
+The system SHALL provide a `ShortcutsPlugin` (an `EditorjsPlugin`) that maps keyboard shortcuts declared in a tool's `options.shortcut` to inline-tool application through the `EditorAPI`.
+
+#### Scenario: Triggering an inline tool via shortcut
+- **GIVEN** an inline tool is registered with `options.shortcut` set to a key combination (e.g. `CMD+B`)
+- **WHEN** that key combination is pressed while the editor has focus
+- **THEN** `ShortcutsPlugin` applies the corresponding inline tool to the current selection via the `EditorAPI`
+
+Implemented in `src/plugins/ShortcutsPlugin.ts`.
diff --git a/openspec/specs/dom-adapters/spec.md b/openspec/specs/dom-adapters/spec.md
new file mode 100644
index 00000000..866ab459
--- /dev/null
+++ b/openspec/specs/dom-adapters/spec.md
@@ -0,0 +1,82 @@
+# DOM Adapters
+
+## Purpose
+
+`@editorjs/dom-adapters` is the glue layer between the document model and real browser DOM elements. It has no rendering/UI logic of its own: it listens to model change events (text added/removed, formatted/unformatted, caret updates) and mutates the DOM to match, and listens to native DOM events (`beforeinput`, `selectionchange`) and translates them into model API calls. It is exposed as an `EditorJSAdapterPlugin` (`DOMAdapters`) wired through an Inversify DI container.
+
+## Requirements
+
+### Requirement: DOMAdapters plugin entry point
+The system SHALL provide `DOMAdapters` as the `EditorJSAdapterPlugin` entry point that builds the DI container, creates/destroys a `DOMBlockToolAdapter` per block, and instantiates the singleton `FormattingAdapter` and `CaretAdapter`.
+
+#### Scenario: Creating a block adapter
+- **GIVEN** a new block is added to the document
+- **WHEN** `DOMAdapters.createBlockToolAdapter` is invoked for that block
+- **THEN** a `DOMBlockToolAdapter` scoped to that block is created and returned via the DI container
+
+Implemented in `src/index.ts`, `src/tokens.ts`.
+
+### Requirement: Shared input registry
+The system SHALL maintain an `InputsRegistry` singleton mapping `(blockId, dataKey)` to the corresponding `HTMLElement`, as the single source of truth queried and mutated by both `DOMBlockToolAdapter` and `CaretAdapter`.
+
+#### Scenario: Looking up a bound input element
+- **GIVEN** a tool has previously bound an `HTMLElement` to a `(blockId, dataKey)` pair via the registry
+- **WHEN** `CaretAdapter` or `DOMBlockToolAdapter` looks up that pair
+- **THEN** the same registered `HTMLElement` is returned
+
+Implemented in `src/InputsRegistry/index.ts`.
+
+### Requirement: Block-level DOM binding and input translation
+The system SHALL provide `DOMBlockToolAdapter`, attached by tools to a contenteditable element for a model data key, which intercepts `beforeinput` events and applies model text-add/remove events back into the DOM.
+
+#### Scenario: Text insertion via beforeinput
+- **GIVEN** a `beforeinput` event of type `insertText`, `insertCompositionText`, `insertFromPaste`, `insertFromDrop`, or `insertReplacementText`
+- **WHEN** it fires on a bound input
+- **THEN** the adapter computes the absolute offset and calls `api.text.insert(...)`, letting the model — not the browser — own the DOM mutation
+
+#### Scenario: Deleting across a cross-input selection
+- **GIVEN** a non-collapsed selection where the input is the "middle" block of a multi-block selection
+- **WHEN** a delete `beforeinput` event fires
+- **THEN** the adapter calls `api.blocks.delete` (whole-block delete) instead of removing a text range
+
+#### Scenario: Splitting a block on Enter
+- **GIVEN** the caret is within a bound input
+- **WHEN** an `insertParagraph` `beforeinput` event fires
+- **THEN** any selected text is removed, `api.blocks.split` is called, and after the next animation frame the caret moves to offset `[0,0]` of the newly created next block
+
+#### Scenario: Model-to-DOM text sync
+- **GIVEN** a `TextAddedEvent` or `TextRemovedEvent` affects a bound input's data key
+- **WHEN** the event is received
+- **THEN** the adapter inserts/deletes DOM text nodes at the mapped boundary point, normalizes the node, and updates the caret index to the new position
+
+Implemented in `src/BlockToolAdapter/index.ts`, `src/BlockToolAdapter/types/InputType.ts`.
+
+### Requirement: Caret and selection synchronization
+The system SHALL provide `CaretAdapter`, which tracks `document.selectionchange`, converts DOM `Selection`/`Range` into a model `Index` (including composite indexes for cross-input selections), and mirrors model caret updates back onto the DOM selection for both native inputs and contenteditable elements.
+
+#### Scenario: Classifying a multi-block selection
+- **GIVEN** a browser selection spanning multiple bound inputs
+- **WHEN** the adapter maps it to indexes
+- **THEN** each input's portion is classified as containing the whole selection, only its start, only its end, or being fully "in between," and segments are sorted into document order before being combined into a composite `Index`
+
+#### Scenario: Avoiding redundant DOM selection updates
+- **GIVEN** a model caret update whose resulting selection is identical to the current DOM selection
+- **WHEN** the adapter would otherwise restore `selectionStart`/`selectionEnd` (native input) or a `Range`/`Selection` (contenteditable)
+- **THEN** it skips reapplying the identical selection to avoid interrupting native browser behavior
+
+Implemented in `src/CaretAdapter/index.ts`, `src/utils/selectionRangeInInput.ts` (validated by `src/utils/selectionRangeInInput.spec.ts`), `src/utils/getAbsoluteRangeOffset.ts`, `getRelativeIndex.ts`.
+
+#### Scenario: Range clipping against an input's boundaries
+- **GIVEN** a selection range and a specific input element
+- **WHEN** `getClippedTextRangeForInput` is computed
+- **THEN** it returns `null` if the range doesn't intersect the input, `[0, textLength]` if the input sits fully between two other blocks' selection anchors, and a normalized `[start, end]` if either anchor falls inside the input
+
+### Requirement: Inline formatting rendering
+The system SHALL provide `FormattingAdapter`, which registers `InlineTool` instances (from `core:ToolLoaded` events) and renders `InlineFragment`s as DOM wrapper elements, re-rendering affected ranges on `TextFormattedEvent`/`TextUnformattedEvent`.
+
+#### Scenario: Re-rendering overlapping formats
+- **GIVEN** a `TextFormattedEvent` or `TextUnformattedEvent` affecting a range
+- **WHEN** the adapter re-renders that range
+- **THEN** it expands the range by one character on each side to catch adjacent boundary fragments, extracts the range into a template, wraps overlapping fragments using each tool's `createWrapper(data)` output, and replaces the DOM range contents so overlapping/adjacent inline formats nest correctly
+
+Implemented in `src/FormattingAdapter/index.ts`, `src/utils/surround.ts`, `src/utils/expandRangeNodeBoundary.ts`.
diff --git a/openspec/specs/model-types/spec.md b/openspec/specs/model-types/spec.md
new file mode 100644
index 00000000..0ad4bb4a
--- /dev/null
+++ b/openspec/specs/model-types/spec.md
@@ -0,0 +1,70 @@
+# Model Types
+
+## Purpose
+
+`@editorjs/model-types` provides the shared, dependency-free low-level types, nominal identifiers, event payload shapes, and base event classes used internally by `@editorjs/model` and `@editorjs/sdk`. It contains no editor runtime logic beyond the `Index`/`IndexBuilder` location-pointer utilities, `keypath` object helpers, and block ID generation. It exists so `model` and `sdk` can share type definitions without depending on each other, and is not intended for direct use by other packages or tools.
+
+## Requirements
+
+### Requirement: Nominal identifier types
+The system SHALL provide nominal (branded) types for `BlockId`, `BlockToolName`, `BlockTuneName`, `DataKey`, and `InlineToolName` so that structurally identical primitives (e.g. two different kinds of string IDs) are not interchangeable at the type level.
+
+#### Scenario: Generating a block id
+- **GIVEN** a new block is being created
+- **WHEN** `generateBlockId()` is called
+- **THEN** it returns a 21-character URL-safe random `BlockId`, generated via `crypto.getRandomValues`
+
+Implemented in `src/Nominal.ts`, `src/BlockId.ts`, `src/BlockTune.ts`, `src/DataKey.ts`, `src/InlineTool.ts`.
+
+### Requirement: Document and block serialization shapes
+The system SHALL define the serialized shapes for blocks, text nodes, and documents (`BlockData`, `BlockNodeInit`, `BlockNodeSerialized`, `TextNodeSerialized`, `EditorDocumentSerialized`, `ChangeData`) so `model` and `sdk` agree on wire/storage format.
+
+#### Scenario: Distinguishing child node types
+- **GIVEN** a block's data tree contains both plain values and text nodes
+- **WHEN** a data node is serialized
+- **THEN** it is tagged with the hidden discriminator property `$t` (`NODE_TYPE_HIDDEN_PROP`) set to a `BlockChildType` (`Value` or `Text`) so consumers can distinguish node kinds without inspecting shape
+
+Implemented in `src/BlockNode.ts`, `src/Text.ts`, `src/Value.ts`, `src/EditorDocument.ts`, `src/ChangeData.ts`, `src/BlockChildType.ts`.
+
+### Requirement: Index — composite document location pointer
+The system SHALL provide an `Index` class representing a location within a document (document id, block index, data key, tune name/key, property name, and/or text range, or a composite of multiple text indexes), with validation that rejects structurally inconsistent combinations.
+
+#### Scenario: Invalid combination of index fields
+- **GIVEN** an `Index` is being validated
+- **WHEN** `tuneName` is set without `tuneKey`, or `tuneName` is combined with `dataKey`/`textRange`, or `propertyName` is combined with any block-related field, or `blockIndex`+`textRange` is set without `dataKey`, or `documentId` is combined with `dataKey`/`tuneName`/`tuneKey`/`textRange` without `blockIndex`
+- **THEN** `validate()` throws an `Invalid index` error
+
+#### Scenario: Composite index for cross-block selections
+- **GIVEN** a selection spans multiple text ranges (e.g. across blocks)
+- **WHEN** an `Index` is built with `compositeSegments`
+- **THEN** it SHALL contain at least two segments, each of which is itself a valid text index (`isTextIndex === true`), and no other root-level index fields may be set simultaneously; an empty `compositeSegments` array is treated as a non-composite (legacy) index
+
+#### Scenario: Parsing a serialized index
+- **GIVEN** a serialized index value
+- **WHEN** `Index.parse` is called
+- **THEN** it accepts either a plain string (legacy single-index form) or an object with a `composite` array of string segments, and throws otherwise
+
+Implemented in `src/Index/Index.ts`, `src/Index/IndexBuilder.ts`, validated by `src/Index/Index.spec.ts`.
+
+### Requirement: Base document event and event bus
+The system SHALL provide a `BaseDocumentEvent` base class carrying `{ index, action, data, userId }` that fires under the `EventType.Changed` DOM event type, concrete event subclasses for each kind of document mutation (block/data/text add/remove/modify, tune/property modification), and a thin `EventBus` (`EventTarget` subclass) used to dispatch them.
+
+#### Scenario: Emitting a document mutation event
+- **GIVEN** a block, text, or value node is added, removed, or modified
+- **THEN** the corresponding concrete event (e.g. `DataNodeAddedEvent`, `TextFormattedEvent`, `BlockAddedEvent`, `PropertyModifiedEvent`, `TuneModifiedEvent`) is dispatched as `EventType.Changed`, carrying an `EventAction` (`Added`/`Removed`/`Modified`) and the affected `Index`
+
+#### Scenario: Emitting a caret event
+- **GIVEN** a caret is added, removed, or updated
+- **THEN** the corresponding caret event (`CaretManagerCaretAddedEvent`, `CaretManagerCaretRemovedEvent`, `CaretManagerCaretUpdatedEvent`) is dispatched as `EventType.CaretManagerUpdated`, carrying the serialized `Caret`
+
+Implemented in `src/BaseDocumentEvent.ts`, `src/EventBus.ts`, `src/EventType.ts`, `src/EventAction.ts`, `src/EventMap.ts`, `src/events/*`.
+
+### Requirement: Keypath utilities
+The system SHALL provide generic dot-path object utilities (`get`, `set`, `has`, `insert`, `remove`, `renumberKeys`) for reading and mutating nested data by string path.
+
+#### Scenario: Reading and mutating by path
+- **GIVEN** a nested object and a dot-separated path string
+- **WHEN** `get`/`set`/`has`/`insert`/`remove` is called with that path
+- **THEN** the corresponding nested value is read, written, checked, inserted, or removed, including array paths (e.g. `items.0.text`)
+
+Implemented in `src/keypath.ts`, validated by `src/keypath.spec.ts`. (Noted in `src/index.ts` as a candidate for extraction into a standalone `@editorjs/utils` package.)
diff --git a/openspec/specs/model/spec.md b/openspec/specs/model/spec.md
new file mode 100644
index 00000000..b156dd4e
--- /dev/null
+++ b/openspec/specs/model/spec.md
@@ -0,0 +1,112 @@
+# Model
+
+## Purpose
+
+`@editorjs/model` is the in-memory document model engine: a tree of blocks containing text/value data nodes and inline-formatting fragments, plus multi-user caret tracking. It centralizes CRUD on document structure, text content, and inline formatting, emitting typed events (from `@editorjs/model-types`) for every mutation so collaboration and rendering layers can react. It is an internal engine consumed by `core` and `ot-server`; tools and plugins should depend on `@editorjs/sdk` instead.
+
+## Requirements
+
+### Requirement: EditorJSModel public facade
+The system SHALL expose `EditorJSModel` as the single entry point wrapping `EditorDocument` and `CaretManager`, providing block/data/text/caret CRUD, tagging mutations with the acting `_userId`, and re-dispatching document events as `ModelEvents`.
+
+#### Scenario: Local caret follows own edits
+- **GIVEN** a user's caret is positioned within a text range
+- **WHEN** that same user inserts or removes text before the caret
+- **THEN** the caret's position updates to remain consistent with the new text length
+
+#### Scenario: Remote text edits shift local caret
+- **GIVEN** a remote user's `TextAddedEvent` or `TextRemovedEvent` affects a block/data key at or before the local caret's `textRange`
+- **WHEN** the event is applied to the model
+- **THEN** the local caret's `textRange` shifts by the inserted/removed length; if the change occurs after the caret's position, the caret is left unchanged
+
+#### Scenario: Remote block structure changes shift local caret
+- **GIVEN** a remote `BlockAddedEvent` or `BlockRemovedEvent` occurs at a block index less than or equal to the local caret's `blockIndex`
+- **WHEN** the event is applied to the model
+- **THEN** the local caret's `blockIndex` shifts by +1 or -1 accordingly; otherwise it is left unchanged
+
+Implemented in `src/EditorJSModel.ts`, validated by `src/EditorJSModel.spec.ts` and `src/EditorJSModel.integration.spec.ts`.
+
+### Requirement: EditorDocument block container
+The system SHALL provide `EditorDocument` as the root container of `BlockNode`s, supporting add/remove/get by index or id (with an O(1) id lookup map), document-level properties, and generic index-based `insertData`/`removeData`/`modifyData` dispatch, bubbling child events with the block index injected.
+
+#### Scenario: Adding a block at an out-of-bounds index
+- **GIVEN** a document with N blocks
+- **WHEN** `addBlock` is called with an index less than 0 or greater than N
+- **THEN** it throws an "Index out of bounds" error
+
+#### Scenario: Adding a block with a duplicate id
+- **GIVEN** a block id that already exists in the document
+- **WHEN** `addBlock` is called reusing that id
+- **THEN** it throws `BlockAlreadyExistsError`
+
+#### Scenario: Appending a block with no index
+- **GIVEN** an existing document
+- **WHEN** `addBlock` is called without an index
+- **THEN** the block is appended at the end and a `BlockAddedEvent` fires carrying the serialized block data and its index
+
+Implemented in `src/entities/EditorDocument/index.ts`, validated by its co-located `.spec.ts`.
+
+### Requirement: BlockNode data tree
+The system SHALL provide `BlockNode` representing a single block's data tree (nested value/text nodes, arrays, and objects) plus its `BlockTune`s, supporting creation/removal/lookup of data nodes, text insert/remove/format/unformat, tune/value updates, and text content extraction.
+
+#### Scenario: Creating a data node at an existing key
+- **GIVEN** a block already has a data node at a given key
+- **WHEN** `createDataNode` is called for that same key
+- **THEN** it throws `AlreadyExistingKeyError`
+
+#### Scenario: Creating a data node within an array path
+- **GIVEN** a block has an array of data nodes at a path
+- **WHEN** `createDataNode` targets an index within that array
+- **THEN** existing entries at or after that index are shifted rather than the operation erroring
+
+Implemented in `src/entities/BlockNode/index.ts`, validated by its co-located `.spec.ts`.
+
+### Requirement: Inline text tree
+The system SHALL provide `TextNode` (extending `ParentInlineNode`) as the inline-fragment tree root for a text data node, supporting text insert/remove, format/unformat over ranges, fragment listing, and index/range validation with auto-merging/normalization of children.
+
+#### Scenario: Inserting text into an empty tree
+- **GIVEN** a `TextNode` with no children
+- **WHEN** text is inserted
+- **THEN** a child inline text node is auto-created and a `TextAddedEvent` is emitted
+
+#### Scenario: Removing text with an out-of-range index
+- **GIVEN** a `TextNode` with a bounded length
+- **WHEN** `removeText` is called with indices outside that range
+- **THEN** it throws a range/index validation error
+
+#### Scenario: Overlapping formatting from the same or different tools
+- **GIVEN** a text range that already has formatting applied
+- **WHEN** `format` is applied again over an overlapping range (same tool or a different tool)
+- **THEN** the resulting fragments nest or merge correctly without duplicating formatting
+
+Implemented in `src/entities/inline-fragments/ParentInlineNode/index.ts`, `src/entities/inline-fragments/TextNode/index.ts`, validated by `src/entities/inline-fragments/TextNode/*.spec.ts` and `src/entities/inline-fragments/specs/InlineTree.integration.spec.ts`.
+
+### Requirement: Value nodes and block tunes
+The system SHALL provide `ValueNode` for leaf non-text data storage and `BlockTune` for per-block tune data, both participating in the block's event-emitting data tree.
+
+#### Scenario: Modifying a value node
+- **GIVEN** a `ValueNode` holding a leaf value within a block
+- **WHEN** its value is updated
+- **THEN** a `ValueModifiedEvent` is emitted carrying the previous and new value
+
+Implemented in `src/entities/ValueNode`, `src/entities/BlockTune`.
+
+### Requirement: Caret management
+The system SHALL provide `CaretManager` as a per-user-id registry of `Caret`s (create/get/update/remove, dispatching Added/Updated/Removed events) and `Caret` as a holder of a nullable `Index` whose `update()` dispatches a `CaretUpdatedEvent` and which serializes via `toJSON()`.
+
+#### Scenario: Registering a new user's caret
+- **GIVEN** a user with no existing caret in the document
+- **WHEN** `CaretManager` creates a caret for that user id
+- **THEN** a `CaretManagerCaretAddedEvent` is dispatched carrying the serialized caret
+
+Implemented in `src/CaretManagement/CaretManager.ts`, `src/CaretManagement/Caret/Caret.ts`, validated by their co-located `.spec.ts` files.
+
+### Requirement: Tool registry and IoC container
+The system SHALL provide an internal `ToolsRegistry` and IoC container for tool/service lookup used by the model engine.
+
+#### Scenario: Resolving a registered tool
+- **GIVEN** a tool has been registered with the model's IoC container
+- **WHEN** internal model code resolves that tool by its token
+- **THEN** the registered instance is returned
+
+Implemented in `src/tools/ToolsRegistry.ts`, `src/IoC/`.
diff --git a/openspec/specs/ot-server/spec.md b/openspec/specs/ot-server/spec.md
new file mode 100644
index 00000000..976ab85b
--- /dev/null
+++ b/openspec/specs/ot-server/spec.md
@@ -0,0 +1,97 @@
+# OT Server
+
+## Purpose
+
+`@editorjs/ot-server` is a standalone Node.js WebSocket server implementing operational transformation for real-time collaborative editing. It accepts client connections keyed by document id, holds authoritative per-document state using `@editorjs/model`'s `EditorJSModel`, transforms and serializes incoming operations from `@editorjs/collaboration-manager`, and broadcasts transformed operations to all other clients editing the same document. It has no persistence layer — all state is in-memory and dropped when the last client for a document disconnects.
+
+## Requirements
+
+### Requirement: WebSocket connection and handshake
+The system SHALL listen on `process.env.WSS_PORT` (default `8080`) and, on receiving a `Handshake` message (`{ document, data? }`), create a `DocumentManager` for that document id on first connection (optionally seeding it via `initializeDocument(data)`), register the socket against that document, and reply with `{ type: Handshake, payload: { ...payload, rev, data } }` — including current server `data` only for clients joining after the first.
+
+#### Scenario: First client seeds the document
+- **GIVEN** no client has yet connected for a given document id
+- **WHEN** the first client sends a `Handshake` with `data`
+- **THEN** the server creates a `DocumentManager` seeded from that `data`, and does not echo `data` back in its handshake reply
+
+#### Scenario: Late joiner receives current state
+- **GIVEN** a `DocumentManager` already exists for a document id
+- **WHEN** a subsequent client sends a `Handshake` for that document
+- **THEN** the server replies with the current server-side document state and revision
+
+Implemented in `src/OTServer.ts`.
+
+### Requirement: Operation processing and broadcast
+The system SHALL, on receiving an `Operation` message, rehydrate it via `Operation.from(payload)`, resolve the `DocumentManager` for `operation.index.documentId`, call `manager.process(operation)`, and on success broadcast `{ type: Operation, payload: processedOperation.serialize() }` to every client subscribed to that document, including the sender.
+
+#### Scenario: Broadcasting a successfully processed operation
+- **GIVEN** a client sends a valid `Operation` for an existing document
+- **WHEN** `DocumentManager.process()` resolves successfully
+- **THEN** the transformed operation is serialized and sent to all clients of that document, including the originating client
+
+Implemented in `src/OTServer.ts`, `src/DocumentManager.ts`.
+
+### Requirement: Protocol error handling
+The system SHALL close the client socket with code `4400` (`BAD_REQUEST_CODE`) and a descriptive reason when a handshake or operation message is missing a `documentId`, references an unknown document, or is rejected by the `DocumentManager`.
+
+#### Scenario: Missing documentId
+- **GIVEN** a client sends a `Handshake` or `Operation` message with no `documentId`
+- **WHEN** the server processes the message
+- **THEN** it closes the socket with code `4400` and a reason describing the missing field
+
+Implemented in `src/OTServer.ts`.
+
+### Requirement: Connection lifecycle and state teardown
+The system SHALL remove a disconnecting socket from its document's client set, and when that was the last client for a document, delete both the client set and the `DocumentManager` for that document, discarding all in-memory state.
+
+#### Scenario: Last client disconnects
+- **GIVEN** exactly one client remains connected for a document
+- **WHEN** that client's socket closes
+- **THEN** the server deletes both the client set and the `DocumentManager` for that document, discarding its in-memory state
+
+Implemented in `src/OTServer.ts` (untested — no test exercises the connect/disconnect lifecycle directly).
+
+### Requirement: Operation transformation against history
+The system SHALL apply incoming operations to the per-document `EditorJSModel` via `DocumentManager`, serializing concurrent `process()` calls through an internal promise chain, and transforming any operation whose revision is behind the current history against all operations submitted since that revision.
+
+#### Scenario: Sequential same-user operations
+- **GIVEN** three consecutive `Insert` operations from the same user with correctly incrementing `rev` values
+- **WHEN** each is processed in order
+- **THEN** they are applied in order, producing the concatenated result
+
+#### Scenario: Concurrent operations at the same revision
+- **GIVEN** two `Insert` operations both submitted at the same revision
+- **WHEN** both are processed
+- **THEN** they are transformed against each other rather than one being rejected — the second is shifted to insert after the first
+
+#### Scenario: Late-arriving older operation
+- **GIVEN** an operation submitted with an older revision arrives after a newer operation has already been processed
+- **WHEN** it is processed
+- **THEN** it is transformed against all operations in the history with revision greater than or equal to its own, producing a correctly ordered result
+
+#### Scenario: Overlapping process() calls stay ordered
+- **GIVEN** multiple `process()` calls are fired without awaiting each other
+- **WHEN** they are handled internally
+- **THEN** they are serialized via promise chaining (`#operationInProcessing`) so operations still apply in submission order deterministically
+
+#### Scenario: Rejecting an operation ahead of server revision
+- **GIVEN** an operation whose `rev` is greater than the server's `#currentRev`
+- **WHEN** `#processNextOperation` evaluates it
+- **THEN** processing resolves to `null` and the error is logged, which `OTServer` surfaces as closing the socket with code `4400`
+
+#### Scenario: Document initialization
+- **GIVEN** a `DocumentManager` created for a new document
+- **WHEN** `initializeDocument(data)` is called with the first client's handshake data
+- **THEN** the internal `EditorJSModel` is seeded from `data`, and `currentModelState()` returns its serialized form for late-joining clients
+
+Implemented in `src/DocumentManager.ts`, validated by `src/DocumentManager.spec.ts`.
+
+### Requirement: Server bootstrap
+The system SHALL load environment configuration via `dotenv`, optionally initialize Hawk error tracking in production, and start `OTServer`.
+
+#### Scenario: Starting in production
+- **GIVEN** the process is started with a production environment configuration
+- **WHEN** `src/index.ts` runs
+- **THEN** environment variables are loaded via `dotenv`, Hawk error tracking is initialized, and `OTServer` is started listening on `WSS_PORT`
+
+Implemented in `src/index.ts`.
diff --git a/openspec/specs/sdk/spec.md b/openspec/specs/sdk/spec.md
new file mode 100644
index 00000000..da3c963a
--- /dev/null
+++ b/openspec/specs/sdk/spec.md
@@ -0,0 +1,89 @@
+# SDK
+
+## Purpose
+
+`@editorjs/sdk` is the shared-contracts package that tools, plugins, and adapters depend on. It provides type-only interfaces for block tools, inline tools, block tunes, and plugins; the concrete `BlockToolAdapter` base class bridging model events to per-block DOM adapters; the `EventBus`/typed event class hierarchy; and the `EditorAPI` surface passed into tools and plugins. It has no dependency on `@editorjs/core` or `@editorjs/model` directly, letting tool authors depend only on this package.
+
+## Requirements
+
+### Requirement: Typed EventBus for channel:name events
+The system SHALL re-export `EventBus` (a thin `EventTarget` subclass defined in `@editorjs/model-types`) and augment `EventTarget`/`addEventListener`/`removeEventListener` typings so listeners are typed by a `` `${Channel}:${Name}` `` event-type template.
+
+#### Scenario: Subscribing to a typed core event
+- **GIVEN** a consumer holds a reference to an `EventBus` instance
+- **WHEN** it calls `addEventListener('core:block:added', handler)`
+- **THEN** the handler is typed to receive the corresponding `CoreEventBase` subclass (e.g. `BlockAddedCoreEvent`) with its specific payload type
+
+Since `EventBus` itself carries no custom subscribe/unsubscribe/error-handling logic, its dispatch semantics (multiple listeners, synchronous order, no built-in error isolation between listeners) are exactly native `EventTarget`/`CustomEvent` semantics.
+
+Implemented in `src/entities/EventBus/EventBus.ts`, `Event.ts`, `events/core/*`, `events/ui/*`, `events/adapter/*` (base classes `CoreEventBase`, `UIEventBase`; enums `CoreEventType`, `AdapterEventType`).
+
+### Requirement: Tool and tune contracts
+The system SHALL define the static/instance contracts that block tools, inline tools, and block tunes must satisfy: `BaseTool`/`BaseToolConstructor` (common `name`, `options`, `prepare()`, `reset()`), `BlockTool`/`BlockToolConstructor` (adds `toolbox`, `shortcut`, `inlineToolbar`, `tunes`, `conversionConfig`, `canBeSplit`), `InlineTool`/`InlineToolConstructor` (adds `isActive`, `getFormattingOptions`, `createWrapper`, `getToolbarConfig`), and `BlockTune`/`BlockTuneConstructor`.
+
+#### Scenario: Options and config merging in a tool facade
+- **GIVEN** a tool class has static `options` (and optionally `options.config`)
+- **WHEN** the tool is registered via `use(Tool, options)` with overriding options
+- **THEN** the facade's `options` getter merges static options with `use()`-time options, with `use()`-time keys taking precedence, and the `config` getter merges similarly, injecting `defaultPlaceholder` only when `isDefault` is true and no `placeholder` key is already present
+
+#### Scenario: Text content conversion without config
+- **GIVEN** a block tool has no `conversionConfig`
+- **WHEN** `exportTextContent`/`importTextContent` is called on its facade
+- **THEN** it throws a descriptive error stating the tool does not have export/import configuration
+
+#### Scenario: Text content conversion with a keypath
+- **GIVEN** a block tool's `conversionConfig` specifies a dot-notation string key (including nested array paths, e.g. `items.0.text`)
+- **WHEN** `exportTextContent`/`importTextContent` is called
+- **THEN** the value at that keypath is read/written, producing or consuming a `TextNodeSerialized` value tagged with the hidden `Text` block-child-type marker
+
+Implemented in `src/entities/BaseTool.ts`, `BlockTool.ts`, `InlineTool.ts`, `BlockTune.ts`, `src/tools/facades/{BaseToolFacade,BlockToolFacade,InlineToolFacade,BlockTuneFacade}.ts`, validated by `src/tools/facades/BaseToolFacade.spec.ts`.
+
+### Requirement: Tools collection
+The system SHALL provide a `ToolsCollection` (a `Map<string, ToolFacadeClass>` subclass) exposing filtered `blockTools`/`inlineTools`/`blockTunes` views over registered tool facades.
+
+#### Scenario: Filtering registered facades by kind
+- **GIVEN** a `ToolsCollection` containing a mix of block tool, inline tool, and tune facades
+- **WHEN** `blockTools`, `inlineTools`, or `blockTunes` is accessed
+- **THEN** each getter returns only the facades matching that kind
+
+Implemented in `src/tools/ToolsCollection.ts`.
+
+### Requirement: BlockToolAdapter bridges model events to per-block DOM state
+The system SHALL provide an abstract `BlockToolAdapter` (extending `EventTarget`) that subscribes to `api.document.onUpdate`, translates model-level `DataNodeAddedEvent`/`DataNodeRemovedEvent`/`ValueModifiedEvent` into adapter-level `KeyAddedEvent`/`KeyRemovedEvent`/`ValueNodeChangedEvent` dispatched on itself, and exposes `registerTextInputKey`, `registerValueKey`, `removeKey`, and `destroy()`.
+
+#### Scenario: Cleaning up an adapter
+- **GIVEN** a `BlockToolAdapter` instance is subscribed to model updates
+- **WHEN** `destroy()` is called
+- **THEN** it unsubscribes its model listener so no further translated events are dispatched
+
+Implemented in `src/entities/BlockToolAdapter.ts`.
+
+### Requirement: Plugin contracts
+The system SHALL define `EditorjsPlugin`/`EditorjsPluginConstructor` (generic UI-plugin contract with optional `destroy()` and a static `type`) and `EditorJSAdapterPlugin`/`EditorjsAdapterPluginConstructor` (singleton adapter plugin contract with `createBlockToolAdapter`/`destroyBlockToolAdapter`).
+
+#### Scenario: Destroying a plugin
+- **GIVEN** a registered `EditorjsPlugin` implements an optional `destroy()` method
+- **WHEN** the editor tears down
+- **THEN** `destroy()` is called on the plugin instance to release its resources
+
+Implemented in `src/entities/EditorjsPlugin.ts`, `src/entities/EditorjsAdapterPlugin.ts`, `src/entities/EntityType.ts`.
+
+### Requirement: EditorAPI surface
+The system SHALL expose an `EditorAPI` type aggregating `BlocksAPI`, `SelectionAPI`, `DocumentAPI`, and `TextAPI` — the API object passed into tools, plugins, and adapters.
+
+#### Scenario: Tool receives the aggregated API
+- **GIVEN** a tool is constructed by the core orchestrator
+- **WHEN** its constructor options are built
+- **THEN** it receives a single `EditorAPI` object exposing `blocks`, `selection`, `document`, and `text` sub-APIs
+
+Implemented in `src/api/EditorAPI.ts`, `src/api/{BlocksAPI,DocumentAPI,SelectionAPI,TextAPI}.ts`.
+
+### Requirement: Keyboard shortcut matching and keypath utilities
+The system SHALL provide `matchKeyboardShortcut(event, shortcutString)` to test a `KeyboardEvent` against a shortcut string (e.g. `CMD+SHIFT+H`), and re-export the keypath utilities (`get`/`set`/`has`/`insert`/`remove`/`renumberKeys`) from `@editorjs/model-types`.
+
+#### Scenario: Matching a shortcut string
+- **GIVEN** a `KeyboardEvent` and a shortcut string such as `CMD+SHIFT+H`
+- **WHEN** `matchKeyboardShortcut(event, 'CMD+SHIFT+H')` is called
+- **THEN** it returns `true` only if the event's modifier keys and key code match the shortcut string exactly
+
+Implemented in `src/utils/keyboardShortcut.ts`, `src/utils/keypath.ts`.
diff --git a/openspec/specs/tools/spec.md b/openspec/specs/tools/spec.md
new file mode 100644
index 00000000..313cccf7
--- /dev/null
+++ b/openspec/specs/tools/spec.md
@@ -0,0 +1,94 @@
+# Tools
+
+## Purpose
+
+The built-in tool/plugin packages provide the default block and inline tools shipped with the editor: `@editorjs/paragraph` (block), and `@editorjs/bold`, `@editorjs/italic`, `@editorjs/inline-link` (inline). Each implements the `BlockTool`/`InlineTool` contracts from `@editorjs/sdk`.
+
+**Note**: none of these four packages contain test files; the scenarios below are derived directly from source control flow rather than confirmed by dedicated tests.
+
+## Requirements
+
+### Requirement: Paragraph block tool
+The system SHALL provide `Paragraph`, implementing `BlockTool<ParagraphData, ParagraphConfig>`, as the default block tool rendering plain text, with `conversionConfig: { import: 'text', export: 'text' }`.
+
+#### Scenario: Registering the text input
+- **GIVEN** a `Paragraph` instance is constructed with a `DOMBlockToolAdapter`
+- **WHEN** construction completes
+- **THEN** it registers `'text'` as the block's single text input key via `adapter.registerTextInputKey('text')`
+
+#### Scenario: Lazy DOM creation on key added
+- **GIVEN** the adapter dispatches a `KeyAddedEvent` for the `'text'` key
+- **WHEN** `Paragraph` handles the `adapter:updated` event
+- **THEN** it lazily creates a contenteditable `<div>` (cached on subsequent `render()` calls) and calls `adapter.setInput('text', div)`
+
+#### Scenario: Cleanup on key removed
+- **GIVEN** the adapter dispatches a `KeyRemovedEvent` for the `'text'` key
+- **WHEN** `Paragraph` handles the event
+- **THEN** it removes the paragraph element from the DOM and calls `adapter.setInput(key, undefined)` to avoid a duplicate input
+
+Implemented in `packages/tools/paragraph/src/index.ts`.
+
+### Requirement: Bold inline tool
+The system SHALL provide `BoldInlineTool`, implementing `InlineTool` with `name: 'bold'`, `shortcut: 'CMD+B'`, and `intersectType: IntersectType.Extend`, rendering a `<b>` wrapper.
+
+#### Scenario: Detecting active state
+- **GIVEN** the current selection range
+- **WHEN** `isActive(range, fragments)` is evaluated
+- **THEN** it returns `true` only if the range is fully contained within an existing bold fragment's range
+
+#### Scenario: Toggling format/unformat
+- **GIVEN** the tool's active state for the current selection
+- **WHEN** `getFormattingOptions(range, fragments)` is called
+- **THEN** it returns `FormattingAction.Unformat` if active, `FormattingAction.Format` otherwise, both carrying the current range
+
+#### Scenario: Extending overlapping bold ranges
+- **GIVEN** `intersectType` is `Extend`
+- **WHEN** a new bold format overlaps an existing bold fragment
+- **THEN** the fragments are merged rather than replaced or split
+
+Implemented in `packages/tools/bold/src/index.ts`.
+
+### Requirement: Italic inline tool
+The system SHALL provide `ItalicInlineTool`, structurally identical to `BoldInlineTool` — `name: 'italic'`, `shortcut: 'CMD+I'`, `intersectType: IntersectType.Extend` — rendering an `<i>` wrapper, with the same containment-based `isActive` and Format/Unformat toggle behavior.
+
+#### Scenario: Toggling italic on a selection
+- **GIVEN** a text selection not currently italicized
+- **WHEN** the italic shortcut (`CMD+I`) or toolbar icon is activated
+- **THEN** `getFormattingOptions` returns `FormattingAction.Format` and the selection is wrapped in `<i>`
+
+Implemented in `packages/tools/italic/src/index.ts`.
+
+### Requirement: Inline link tool
+The system SHALL provide `LinkInlineTool`, implementing `InlineTool` with `name: 'link'` and `intersectType: IntersectType.Replace` (a new link replaces any overlapping link rather than merging), rendering an `<a>` wrapper and exposing a popover for entering/editing the URL.
+
+#### Scenario: Editing an existing link
+- **GIVEN** the current selection is inside an existing link fragment
+- **WHEN** the toolbar renders `getToolbarConfig`
+- **THEN** the icon is `IconUnlink`, the popover pre-opens with the existing `href` value, and activating the icon calls `api.selection.applyInlineTool` with `FormattingAction.Unformat` to remove the link
+
+#### Scenario: Creating a new link
+- **GIVEN** the current selection is not inside an existing link
+- **WHEN** the toolbar renders `getToolbarConfig`
+- **THEN** the popover opens with an empty input, autofocused via a microtask
+
+#### Scenario: Confirming a link URL
+- **GIVEN** the popover's URL input has focus
+- **WHEN** the user presses Enter
+- **THEN** `api.selection.applyInlineTool` is called with `data: { href: <input value> }` and `FormattingAction.Format` (new link) or `FormattingAction.None` (editing an existing link's href), depending on whether the tool was active
+
+#### Scenario: Wrapper href assignment
+- **GIVEN** `createWrapper(data)` is called
+- **WHEN** `data.href` is a string
+- **THEN** the produced `<a>` element's `href` attribute is set to that value; otherwise a bare `<a>` is produced
+
+Implemented in `packages/tools/inline-link/src/index.ts`.
+
+### Requirement: Shared inline-tool activation contract
+The system SHALL have all inline tools (Bold, Italic, Link) implement the same containment-based `isActive`/`getFormattingOptions` pattern, differing only in wrapper tag, keyboard shortcut, and `intersectType`.
+
+#### Scenario: Consistent containment check across tools
+- **GIVEN** any of Bold, Italic, or Link is evaluating `isActive` for the current selection
+- **WHEN** the selection range is fully contained within an existing fragment for that tool
+- **THEN** `isActive` returns `true`, using the same range-containment logic regardless of which tool is evaluating it
+
+Cross-referenced across `packages/tools/bold/src/index.ts`, `packages/tools/italic/src/index.ts`, `packages/tools/inline-link/src/index.ts`, and validated indirectly by the generic facade test `packages/sdk/src/tools/facades/BaseToolFacade.spec.ts`.
diff --git a/openspec/specs/ui/spec.md b/openspec/specs/ui/spec.md
new file mode 100644
index 00000000..c25cab81
--- /dev/null
+++ b/openspec/specs/ui/spec.md
@@ -0,0 +1,94 @@
+# UI
+
+## Purpose
+
+`@editorjs/ui` is the default rendering shell for Editor.js: a set of `EditorjsPlugin` implementations that subscribe to the core `EventBus` to render DOM and dispatch their own `ui:*` events so the pieces can wire themselves together. It owns no document state — it renders what `core` (`BlockManager`/`SelectionManager`, etc.) reports, and forwards user interaction back through `EditorAPI`.
+
+**Note**: this package has no automated test suite (`.spec.ts`/`.test.ts` files); the scenarios below are derived directly from the event-wiring logic in source rather than confirmed by tests.
+
+## Requirements
+
+### Requirement: Shell assembly
+The system SHALL provide `EditorjsUI` as the top-level shell that creates the editor wrapper in the holder element and reactively assembles the Toolbar, InlineToolbar, and Blocks elements into it as each announces its own `*:rendered` event, without holding direct references to those components.
+
+#### Scenario: Assembling the shell from render events
+- **GIVEN** `EditorjsUI` has created the wrapper element
+- **WHEN** `ui:toolbar:rendered`, `ui:inline-toolbar:rendered`, and `ui:blocks:rendered` events are received
+- **THEN** the corresponding rendered elements are appended to the wrapper in the order the events arrive
+
+Implemented in `src/index.ts`.
+
+### Requirement: Blocks holder rendering and input capture
+The system SHALL provide `BlocksUI`, which renders the contenteditable blocks holder, adds/removes block wrappers on `core:BlockAdded`/`core:BlockRemoved`, captures native `beforeinput` and remaps it into a normalized `BeforeInputUIEvent`, handles undo/redo keyboard shortcuts, and dispatches block-hover selection events.
+
+#### Scenario: Inserting a block wrapper at an index
+- **GIVEN** a `BlockAddedCoreEvent` with a valid index
+- **WHEN** `BlocksUI` processes it
+- **THEN** the block element is wrapped and inserted at that position in the blocks holder, or appended if the index is beyond the current list
+
+#### Scenario: Rejecting an invalid block index
+- **GIVEN** a `BlockAddedCoreEvent`/`BlockRemovedCoreEvent` with an out-of-bounds index
+- **WHEN** `BlocksUI` processes it
+- **THEN** it throws an "Index out of bounds" error
+
+#### Scenario: Hovering a block dispatches selection
+- **GIVEN** the pointer enters a rendered block element
+- **WHEN** the `mouseenter` event fires
+- **THEN** `BlocksUI` dispatches a `BlockSelectedUIEvent` carrying the block and its index
+
+#### Scenario: Normalizing beforeinput
+- **GIVEN** a native `beforeinput` event fires on the blocks holder
+- **WHEN** `BlocksUI` intercepts it
+- **THEN** the default action is prevented and a `BeforeInputUIEvent` is dispatched carrying `data`, `inputType`, `isComposing`, and `targetRanges`, distinguishing native-input vs. contenteditable sources and cross-input selections
+
+#### Scenario: Undo/redo keyboard shortcuts
+- **GIVEN** the blocks holder has focus
+- **WHEN** Cmd/Ctrl+Z is pressed
+- **THEN** `api.document.undo()` is called with the default action prevented; if Shift is also held, `api.document.redo()` is called instead
+
+Implemented in `src/Blocks/Blocks.ts`, `src/Blocks/events/*`.
+
+### Requirement: Floating toolbar
+The system SHALL provide `ToolbarUI`, which renders a floating toolbar with a plus-button and actions area, repositions itself to the selected block's offset on `ui:blocks:block-selected` (unless the Toolbox is open), and opens the Toolbox on plus-button click.
+
+#### Scenario: Repositioning on block selection
+- **GIVEN** a `BlockSelectedUIEvent` is received and the Toolbox is not currently open
+- **WHEN** `ToolbarUI` handles the event
+- **THEN** the toolbar moves to the selected block's `offsetTop`
+
+#### Scenario: Opening the toolbox
+- **GIVEN** the user clicks the toolbar's plus button
+- **WHEN** the click is handled
+- **THEN** `ToolbarUI` dispatches a `ToolboxOpenUIEvent`
+
+Implemented in `src/Toolbar/Toolbar.ts`, `src/Toolbar/ToolbarRenderedUIEvent.ts`.
+
+### Requirement: Inline toolbar
+The system SHALL provide `InlineToolbarUI`, which builds a popover of available inline tools on `core:SelectionChanged`, shows/positions/hides itself based on whether the current selection has a non-collapsed text range, and wires tool activation to `api.selection.applyInlineTool`.
+
+#### Scenario: Hiding on collapsed or absent selection
+- **GIVEN** a `SelectionChangedCoreEvent` with no index, no text segments, only a collapsed range, or no browser selection range
+- **WHEN** `InlineToolbarUI` handles the event
+- **THEN** the popover is hidden
+
+#### Scenario: Showing on a non-collapsed selection
+- **GIVEN** a `SelectionChangedCoreEvent` with a non-collapsed text range
+- **WHEN** `InlineToolbarUI` handles the event
+- **THEN** it rebuilds the popover from the available inline tools and current fragments, positions it at the selection's bounding rect, and shows it
+
+Implemented in `src/InlineToolbar/InlineToolbar.ts`, `src/InlineToolbar/InlineToolbarRenderedUIEvent.ts`.
+
+### Requirement: Toolbox
+The system SHALL provide `ToolboxUI`, a searchable popover of block tools populated on `core:ToolLoaded` (only tools where `tool.isBlock()` is true), which opens on `ui:toolbox:open`, tracks the hovered block index to determine insert position, and inserts a new block via `api.blocks.insert` on tool activation.
+
+#### Scenario: Inserting a new block from the toolbox
+- **GIVEN** the toolbox is open and a block index is tracked from the last hover (or none)
+- **WHEN** the user activates a tool entry
+- **THEN** `api.blocks.insert({ type, data, index: trackedIndex + 1 (or undefined), focus: true })` is called, and the popover closes, dispatching `ToolboxClosedUIEvent`
+
+#### Scenario: Toolbox open/close events
+- **GIVEN** the toolbox popover's open state changes
+- **WHEN** it opens or closes
+- **THEN** `ToolboxOpenedUIEvent` or `ToolboxClosedUIEvent` is dispatched accordingly
+
+Implemented in `src/Toolbox/Toolbox.ts`, `src/Toolbox/events/*`, `src/Toolbox/ToolboxConfigEntry.ts`.
diff --git a/packages/.claude/commands/opsx/apply.md b/packages/.claude/commands/opsx/apply.md
new file mode 100644
index 00000000..c6cb9b6d
--- /dev/null
+++ b/packages/.claude/commands/opsx/apply.md
@@ -0,0 +1,156 @@
+---
+name: "OPSX: Apply"
+description: Implement tasks from an OpenSpec change (Experimental)
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! You can archive this change with `/opsx:archive`.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.claude/commands/opsx/archive.md b/packages/.claude/commands/opsx/archive.md
new file mode 100644
index 00000000..df8a2f2d
--- /dev/null
+++ b/packages/.claude/commands/opsx/archive.md
@@ -0,0 +1,161 @@
+---
+name: "OPSX: Archive"
+description: Archive a completed change in the experimental workflow
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, archive, experimental]
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Spec sync status (synced / sync skipped / no delta specs)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success (No Delta Specs)**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** No delta specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success With Warnings**
+
+```
+## Archive Complete (with warnings)
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** Sync skipped (user chose to skip)
+
+**Warnings:**
+- Archived with 2 incomplete artifacts
+- Archived with 3 incomplete tasks
+- Delta spec sync was skipped (user chose to skip)
+
+Review the archive if this was not intentional.
+```
+
+**Output On Error (Archive Exists)**
+
+```
+## Archive Failed
+
+**Change:** <change-name>
+**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+
+Target archive directory already exists.
+
+**Options:**
+1. Rename the existing archive
+2. Delete the existing archive if it's a duplicate
+3. Wait until a different date to archive
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.claude/commands/opsx/explore.md b/packages/.claude/commands/opsx/explore.md
new file mode 100644
index 00000000..7558a631
--- /dev/null
+++ b/packages/.claude/commands/opsx/explore.md
@@ -0,0 +1,175 @@
+---
+name: "OPSX: Explore"
+description: "Enter explore mode - think through ideas, investigate problems, clarify requirements"
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, explore, experimental, thinking]
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A change name: "add-dark-mode" (to explore in context of that change)
+- A comparison: "postgres vs sqlite for this"
+- Nothing (just enter explore mode)
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+If the user mentioned a specific change name, read its artifacts for context.
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.claude/commands/opsx/propose.md b/packages/.claude/commands/opsx/propose.md
new file mode 100644
index 00000000..8d99588f
--- /dev/null
+++ b/packages/.claude/commands/opsx/propose.md
@@ -0,0 +1,110 @@
+---
+name: "OPSX: Propose"
+description: Propose a new change - create it and generate all artifacts in one step
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
+
+**Steps**
+
+1. **If no input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` to start implementing."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.claude/commands/opsx/sync.md b/packages/.claude/commands/opsx/sync.md
new file mode 100644
index 00000000..a3550cf4
--- /dev/null
+++ b/packages/.claude/commands/opsx/sync.md
@@ -0,0 +1,144 @@
+---
+name: "OPSX: Sync"
+description: Sync delta specs from a change to main specs
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, specs, experimental]
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/packages/.claude/commands/opsx/update.md b/packages/.claude/commands/opsx/update.md
new file mode 100644
index 00000000..75104f59
--- /dev/null
+++ b/packages/.claude/commands/opsx/update.md
@@ -0,0 +1,82 @@
+---
+name: "OPSX: Update"
+description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, artifacts, experimental]
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.claude/commands/opsx/verify.md b/packages/.claude/commands/opsx/verify.md
new file mode 100644
index 00000000..7f6cade1
--- /dev/null
+++ b/packages/.claude/commands/opsx/verify.md
@@ -0,0 +1,168 @@
+---
+name: "OPSX: Verify"
+description: Verify implementation matches change artifacts before archiving
+allowed-tools: Bash(openspec:*)
+category: Workflow
+tags: [workflow, verify, experimental]
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/packages/.claude/skills/openspec-apply-change/SKILL.md b/packages/.claude/skills/openspec-apply-change/SKILL.md
new file mode 100644
index 00000000..1375861c
--- /dev/null
+++ b/packages/.claude/skills/openspec-apply-change/SKILL.md
@@ -0,0 +1,160 @@
+---
+name: openspec-apply-change
+description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! Ready to archive this change.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.claude/skills/openspec-archive-change/SKILL.md b/packages/.claude/skills/openspec-archive-change/SKILL.md
new file mode 100644
index 00000000..c0c169d1
--- /dev/null
+++ b/packages/.claude/skills/openspec-archive-change/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: openspec-archive-change
+description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Whether specs were synced (if applicable)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
+
+All artifacts complete. All tasks complete.
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use openspec-sync-specs approach (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.claude/skills/openspec-explore/SKILL.md b/packages/.claude/skills/openspec-explore/SKILL.md
new file mode 100644
index 00000000..771271a8
--- /dev/null
+++ b/packages/.claude/skills/openspec-explore/SKILL.md
@@ -0,0 +1,290 @@
+---
+name: openspec-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+      ════════════════════════════════════════════
+
+      Awareness          Coordination         Sync
+          │                   │                 │
+          ▼                   ▼                 ▼
+      ┌────────┐         ┌────────┐        ┌────────┐
+      │Presence│         │Cursors │        │  CRDT  │
+      │  "3    │         │ Multi  │        │Conflict│
+      │online" │         │ select │        │  free  │
+      └────────┘         └────────┘        └────────┘
+          │                   │                 │
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     ┌─────────────────────────────────────────────┐
+     │              CURRENT AUTH FLOW              │
+     └─────────────────────────────────────────────┘
+                        │
+          ┌─────────────┼─────────────┐
+          ▼             ▼             ▼
+     ┌─────────┐  ┌─────────┐  ┌─────────┐
+     │ Google  │  │  GitHub │  │  Email  │
+     │  OAuth  │  │  OAuth  │  │  Magic  │
+     └────┬────┘  └────┬────┘  └────┬────┘
+          │             │             │
+          └─────────────┼─────────────┘
+                        ▼
+                  ┌───────────┐
+                  │  Session  │
+                  └─────┬─────┘
+                        │
+                        ▼
+                  ┌───────────┐
+                  │   Perms   │
+                  └───────────┘
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /opsx:explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads change artifacts]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the design to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     ┌─────────────────────────────────────────────────┐
+     │          CLI TOOL DATA STORAGE                  │
+     └─────────────────────────────────────────────────┘
+
+     Key constraints:
+     • No daemon running
+     • Must work offline
+     • Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ✓      needs server ✗
+     Offline      yes ✓           no ✗
+     Single file  yes ✓           no ✗
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a change proposal
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.claude/skills/openspec-propose/SKILL.md b/packages/.claude/skills/openspec-propose/SKILL.md
new file mode 100644
index 00000000..716d2d3d
--- /dev/null
+++ b/packages/.claude/skills/openspec-propose/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: openspec-propose
+description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
+
+**Steps**
+
+1. **If no clear input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.claude/skills/openspec-sync-specs/SKILL.md b/packages/.claude/skills/openspec-sync-specs/SKILL.md
new file mode 100644
index 00000000..b6aad67a
--- /dev/null
+++ b/packages/.claude/skills/openspec-sync-specs/SKILL.md
@@ -0,0 +1,148 @@
+---
+name: openspec-sync-specs
+description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/packages/.claude/skills/openspec-update-change/SKILL.md b/packages/.claude/skills/openspec-update-change/SKILL.md
new file mode 100644
index 00000000..3e1c931d
--- /dev/null
+++ b/packages/.claude/skills/openspec-update-change/SKILL.md
@@ -0,0 +1,86 @@
+---
+name: openspec-update-change
+description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.claude/skills/openspec-verify-change/SKILL.md b/packages/.claude/skills/openspec-verify-change/SKILL.md
new file mode 100644
index 00000000..b6557c6f
--- /dev/null
+++ b/packages/.claude/skills/openspec-verify-change/SKILL.md
@@ -0,0 +1,172 @@
+---
+name: openspec-verify-change
+description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/packages/.github/prompts/opsx-apply.prompt.md b/packages/.github/prompts/opsx-apply.prompt.md
new file mode 100644
index 00000000..5bd0b6bb
--- /dev/null
+++ b/packages/.github/prompts/opsx-apply.prompt.md
@@ -0,0 +1,152 @@
+---
+description: Implement tasks from an OpenSpec change (Experimental)
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! You can archive this change with `/opsx:archive`.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.github/prompts/opsx-archive.prompt.md b/packages/.github/prompts/opsx-archive.prompt.md
new file mode 100644
index 00000000..f9ba89d8
--- /dev/null
+++ b/packages/.github/prompts/opsx-archive.prompt.md
@@ -0,0 +1,157 @@
+---
+description: Archive a completed change in the experimental workflow
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Prompt user for confirmation to continue
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Spec sync status (synced / sync skipped / no delta specs)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success (No Delta Specs)**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** No delta specs
+
+All artifacts complete. All tasks complete.
+```
+
+**Output On Success With Warnings**
+
+```
+## Archive Complete (with warnings)
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** Sync skipped (user chose to skip)
+
+**Warnings:**
+- Archived with 2 incomplete artifacts
+- Archived with 3 incomplete tasks
+- Delta spec sync was skipped (user chose to skip)
+
+Review the archive if this was not intentional.
+```
+
+**Output On Error (Archive Exists)**
+
+```
+## Archive Failed
+
+**Change:** <change-name>
+**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+
+Target archive directory already exists.
+
+**Options:**
+1. Rename the existing archive
+2. Delete the existing archive if it's a duplicate
+3. Wait until a different date to archive
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.github/prompts/opsx-explore.prompt.md b/packages/.github/prompts/opsx-explore.prompt.md
new file mode 100644
index 00000000..856001a6
--- /dev/null
+++ b/packages/.github/prompts/opsx-explore.prompt.md
@@ -0,0 +1,171 @@
+---
+description: Enter explore mode - think through ideas, investigate problems, clarify requirements
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A change name: "add-dark-mode" (to explore in context of that change)
+- A comparison: "postgres vs sqlite for this"
+- Nothing (just enter explore mode)
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+If the user mentioned a specific change name, read its artifacts for context.
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.github/prompts/opsx-propose.prompt.md b/packages/.github/prompts/opsx-propose.prompt.md
new file mode 100644
index 00000000..fde349a6
--- /dev/null
+++ b/packages/.github/prompts/opsx-propose.prompt.md
@@ -0,0 +1,106 @@
+---
+description: Propose a new change - create it and generate all artifacts in one step
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
+
+**Steps**
+
+1. **If no input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` to start implementing."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.github/prompts/opsx-sync.prompt.md b/packages/.github/prompts/opsx-sync.prompt.md
new file mode 100644
index 00000000..961aa409
--- /dev/null
+++ b/packages/.github/prompts/opsx-sync.prompt.md
@@ -0,0 +1,140 @@
+---
+description: Sync delta specs from a change to main specs
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/packages/.github/prompts/opsx-update.prompt.md b/packages/.github/prompts/opsx-update.prompt.md
new file mode 100644
index 00000000..5b316d4d
--- /dev/null
+++ b/packages/.github/prompts/opsx-update.prompt.md
@@ -0,0 +1,78 @@
+---
+description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.github/prompts/opsx-verify.prompt.md b/packages/.github/prompts/opsx-verify.prompt.md
new file mode 100644
index 00000000..cc1377a7
--- /dev/null
+++ b/packages/.github/prompts/opsx-verify.prompt.md
@@ -0,0 +1,164 @@
+---
+description: Verify implementation matches change artifacts before archiving
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/packages/.github/skills/openspec-apply-change/SKILL.md b/packages/.github/skills/openspec-apply-change/SKILL.md
new file mode 100644
index 00000000..1375861c
--- /dev/null
+++ b/packages/.github/skills/openspec-apply-change/SKILL.md
@@ -0,0 +1,160 @@
+---
+name: openspec-apply-change
+description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Implement tasks from an OpenSpec change.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **Select the change**
+
+   If a name is provided, use it. Otherwise:
+   - Infer from conversation context if the user mentioned a change
+   - Auto-select if only one active change exists
+   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
+
+   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
+   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
+
+3. **Get apply instructions**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns:
+   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
+   - Progress (total, complete, remaining)
+   - Task list with status
+   - Dynamic instruction based on current state
+
+   **Handle states:**
+   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
+   - If `state: "all_done"`: congratulate, suggest archive
+   - Otherwise: proceed to implementation
+
+4. **Read context files**
+
+   Read every file path listed under `contextFiles` from the apply instructions output.
+   The files depend on the schema being used:
+   - **spec-driven**: proposal, specs, design, tasks
+   - Other schemas: follow the contextFiles from CLI output
+
+5. **Show current progress**
+
+   Display:
+   - Schema being used
+   - Progress: "N/M tasks complete"
+   - Remaining tasks overview
+   - Dynamic instruction from CLI
+
+6. **Implement tasks (loop until done or blocked)**
+
+   For each pending task:
+   - Show which task is being worked on
+   - Make the code changes required
+   - Keep changes minimal and focused
+   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
+   - Continue to next task
+
+   **Pause if:**
+   - Task is unclear → ask for clarification
+   - Implementation reveals a design issue → suggest updating artifacts
+   - Error or blocker encountered → report and wait for guidance
+   - User interrupts
+
+7. **On completion or pause, show status**
+
+   Display:
+   - Tasks completed this session
+   - Overall progress: "N/M tasks complete"
+   - If all done: suggest archive
+   - If paused: explain why and wait for guidance
+
+**Output During Implementation**
+
+```
+## Implementing: <change-name> (schema: <schema-name>)
+
+Working on task 3/7: <task description>
+[...implementation happening...]
+✓ Task complete
+
+Working on task 4/7: <task description>
+[...implementation happening...]
+✓ Task complete
+```
+
+**Output On Completion**
+
+```
+## Implementation Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 7/7 tasks complete ✓
+
+### Completed This Session
+- [x] Task 1
+- [x] Task 2
+...
+
+All tasks complete! Ready to archive this change.
+```
+
+**Output On Pause (Issue Encountered)**
+
+```
+## Implementation Paused
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Progress:** 4/7 tasks complete
+
+### Issue Encountered
+<description of the issue>
+
+**Options:**
+1. <option 1>
+2. <option 2>
+3. Other approach
+
+What would you like to do?
+```
+
+**Guardrails**
+- Keep going through tasks until done or blocked
+- Always read context files before starting (from the apply instructions output)
+- If task is ambiguous, pause and ask before implementing
+- If implementation reveals issues, pause and suggest artifact updates
+- Keep code changes minimal and scoped to each task
+- Update task checkbox immediately after completing each task
+- Pause on errors, blockers, or unclear requirements - don't guess
+- Use contextFiles from CLI output, don't assume specific file names
+
+**Fluid Workflow Integration**
+
+This skill supports the "actions on a change" model:
+
+- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
+- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.github/skills/openspec-archive-change/SKILL.md b/packages/.github/skills/openspec-archive-change/SKILL.md
new file mode 100644
index 00000000..c0c169d1
--- /dev/null
+++ b/packages/.github/skills/openspec-archive-change/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: openspec-archive-change
+description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Archive a completed change in the experimental workflow.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show only active changes (not already archived).
+   Include the schema used for each change if available.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check artifact completion status**
+
+   Run `openspec status --change "<name>" --json` to check artifact completion.
+
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - `artifacts`: List of artifacts with their status (`done` or other)
+
+   **If any artifacts are not `done`:**
+   - Display warning listing incomplete artifacts
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+3. **Check task completion status**
+
+   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
+
+   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
+
+   **If incomplete tasks found:**
+   - Display warning showing count of incomplete tasks
+   - Use **AskUserQuestion tool** to confirm user wants to proceed
+   - Proceed if user confirms
+
+   **If no tasks file exists:** Proceed without task-related warning.
+
+4. **Assess delta spec sync state**
+
+   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
+
+   **If delta specs exist:**
+   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
+   - Determine what changes would be applied (adds, modifications, removals, renames)
+   - Show a combined summary before prompting
+
+   **Prompt options:**
+   - If changes needed: "Sync now (recommended)", "Archive without syncing"
+   - If already synced: "Archive now", "Sync anyway", "Cancel"
+
+   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
+
+5. **Perform the archive**
+
+   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
+   ```bash
+   mkdir -p "<planningHome.changesDir>/archive"
+   ```
+
+   Generate target name using current date: `YYYY-MM-DD-<change-name>`
+
+   **Check if target already exists:**
+   - If yes: Fail with error, suggest renaming existing archive or using different date
+   - If no: Move `changeRoot` to the archive directory
+
+   ```bash
+   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
+   ```
+
+6. **Display summary**
+
+   Show archive completion summary including:
+   - Change name
+   - Schema that was used
+   - Archive location
+   - Whether specs were synced (if applicable)
+   - Note about any warnings (incomplete artifacts/tasks)
+
+**Output On Success**
+
+```
+## Archive Complete
+
+**Change:** <change-name>
+**Schema:** <schema-name>
+**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
+**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
+
+All artifacts complete. All tasks complete.
+```
+
+**Guardrails**
+- Always prompt for change selection if not provided
+- Use artifact graph (openspec status --json) for completion checking
+- Don't block archive on warnings - just inform and confirm
+- Preserve .openspec.yaml when moving to archive (it moves with the directory)
+- Show clear summary of what happened
+- If sync is requested, use openspec-sync-specs approach (agent-driven)
+- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.github/skills/openspec-explore/SKILL.md b/packages/.github/skills/openspec-explore/SKILL.md
new file mode 100644
index 00000000..771271a8
--- /dev/null
+++ b/packages/.github/skills/openspec-explore/SKILL.md
@@ -0,0 +1,290 @@
+---
+name: openspec-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
+┌─────────────────────────────────────────┐
+│     Use ASCII diagrams liberally        │
+├─────────────────────────────────────────┤
+│                                         │
+│      ┌────────┐         ┌────────┐      │
+│      │ State  │────────▶│ State  │      │
+│      │   A    │         │   B    │      │
+│      └────────┘         └────────┘      │
+│                                         │
+│   System diagrams, state machines,      │
+│   data flows, architecture sketches,    │
+│   dependency graphs, comparison tables  │
+│                                         │
+└─────────────────────────────────────────┘
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## OpenSpec Awareness
+
+You have full context of the OpenSpec system. Use it naturally, don't force it.
+
+### Check for context
+
+At the start, quickly check what exists:
+```bash
+openspec list --json
+```
+
+This tells you:
+- If there are active changes
+- Their names, schemas, and status
+- What the user might be working on
+
+### When no change exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to start a change. Want me to create a proposal?"
+- Or keep exploring - no pressure to formalize
+
+### When a change exists
+
+If the user mentions a change or you detect one is relevant:
+
+1. **Resolve and read existing artifacts for context**
+   - Run `openspec status --change "<name>" --json`.
+   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
+   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
+
+2. **Reference them naturally in conversation**
+   - "Your design mentions using Redis, but we just realized SQLite fits better..."
+   - "The proposal scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+    | Insight Type               | Where to Capture               |
+    |----------------------------|--------------------------------|
+    | New requirement discovered | `specs/<capability>/spec.md` |
+    | Requirement changed        | `specs/<capability>/spec.md` |
+    | Design decision made       | `design.md`                  |
+    | Scope changed              | `proposal.md`                |
+    | New work identified        | `tasks.md`                   |
+    | Assumption invalidated     | Relevant artifact              |
+
+   Example offers:
+   - "That's a design decision. Capture it in design.md?"
+   - "This is a new requirement. Add it to specs?"
+   - "This changes scope. Update the proposal?"
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+      ════════════════════════════════════════════
+
+      Awareness          Coordination         Sync
+          │                   │                 │
+          ▼                   ▼                 ▼
+      ┌────────┐         ┌────────┐        ┌────────┐
+      │Presence│         │Cursors │        │  CRDT  │
+      │  "3    │         │ Multi  │        │Conflict│
+      │online" │         │ select │        │  free  │
+      └────────┘         └────────┘        └────────┘
+          │                   │                 │
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     ┌─────────────────────────────────────────────┐
+     │              CURRENT AUTH FLOW              │
+     └─────────────────────────────────────────────┘
+                        │
+          ┌─────────────┼─────────────┐
+          ▼             ▼             ▼
+     ┌─────────┐  ┌─────────┐  ┌─────────┐
+     │ Google  │  │  GitHub │  │  Email  │
+     │  OAuth  │  │  OAuth  │  │  Magic  │
+     └────┬────┘  └────┬────┘  └────┬────┘
+          │             │             │
+          └─────────────┼─────────────┘
+                        ▼
+                  ┌───────────┐
+                  │  Session  │
+                  └─────┬─────┘
+                        │
+                        ▼
+                  ┌───────────┐
+                  │   Perms   │
+                  └───────────┘
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /opsx:explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads change artifacts]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the design to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     ┌─────────────────────────────────────────────────┐
+     │          CLI TOOL DATA STORAGE                  │
+     └─────────────────────────────────────────────────┘
+
+     Key constraints:
+     • No daemon running
+     • Must work offline
+     • Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ✓      needs server ✗
+     Offline      yes ✓           no ✗
+     Single file  yes ✓           no ✗
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into a proposal**: "Ready to start? I can create a change proposal."
+- **Result in artifact updates**: "Updated design.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a change proposal
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.github/skills/openspec-propose/SKILL.md b/packages/.github/skills/openspec-propose/SKILL.md
new file mode 100644
index 00000000..716d2d3d
--- /dev/null
+++ b/packages/.github/skills/openspec-propose/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: openspec-propose
+description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Propose a new change - create the change and generate all artifacts in one step.
+
+I'll create a change with artifacts:
+- proposal.md (what & why)
+- design.md (how)
+- tasks.md (implementation steps)
+
+When ready to implement, run /opsx:apply
+
+---
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
+
+**Steps**
+
+1. **If no clear input provided, ask what they want to build**
+
+   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
+   > "What change do you want to work on? Describe what you want to build or fix."
+
+   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
+
+   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
+
+2. **Create the change directory**
+   ```bash
+   openspec new change "<name>"
+   ```
+   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
+
+3. **Get the artifact build order**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to get:
+   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
+   - `artifacts`: list of all artifacts with their status and dependencies
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+4. **Create artifacts in sequence until apply-ready**
+
+   Use the **TodoWrite tool** to track progress through the artifacts.
+
+   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
+
+   a. **For each artifact that is `ready` (dependencies satisfied)**:
+      - Get instructions:
+        ```bash
+        openspec instructions <artifact-id> --change "<name>" --json
+        ```
+      - The instructions JSON includes:
+        - `context`: Project background (constraints for you - do NOT include in output)
+        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
+        - `template`: The structure to use for your output file
+        - `instruction`: Schema-specific guidance for this artifact type
+        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
+        - `dependencies`: Completed artifacts to read for context
+      - Read any completed dependency files for context
+      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
+      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
+      - Show brief progress: "Created <artifact-id>"
+
+   b. **Continue until all `applyRequires` artifacts are complete**
+      - After creating each artifact, re-run `openspec status --change "<name>" --json`
+      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
+      - Stop when all `applyRequires` artifacts are done
+
+   c. **If an artifact requires user input** (unclear context):
+      - Use **AskUserQuestion tool** to clarify
+      - Then continue with creation
+
+5. **Show final status**
+   ```bash
+   openspec status --change "<name>"
+   ```
+
+**Output**
+
+After completing all artifacts, summarize:
+- Change name and location
+- List of artifacts created with brief descriptions
+- What's ready: "All artifacts created! Ready for implementation."
+- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
+
+**Artifact Creation Guidelines**
+
+- Follow the `instruction` field from `openspec instructions` for each artifact type
+- The schema defines what each artifact should contain - follow it
+- Read dependency artifacts for context before creating new ones
+- Use `template` as the structure for your output file - fill in its sections
+- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
+  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
+  - These guide what you write, but should never appear in the output
+
+**Guardrails**
+- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
+- Always read dependency artifacts before creating a new one
+- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
+- If a change with that name already exists, ask if user wants to continue it or create a new one
+- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.github/skills/openspec-sync-specs/SKILL.md b/packages/.github/skills/openspec-sync-specs/SKILL.md
new file mode 100644
index 00000000..b6aad67a
--- /dev/null
+++ b/packages/.github/skills/openspec-sync-specs/SKILL.md
@@ -0,0 +1,148 @@
+---
+name: openspec-sync-specs
+description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Sync delta specs from a change to main specs.
+
+This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have delta specs (under `specs/` directory).
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Resolve change context**
+
+   Run:
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+
+3. **Find delta specs**
+
+   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
+
+   Each delta spec file contains sections like:
+   - `## ADDED Requirements` - New requirements to add
+   - `## MODIFIED Requirements` - Changes to existing requirements
+   - `## REMOVED Requirements` - Requirements to remove
+   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
+
+   If no delta specs found, inform user and stop.
+
+4. **For each delta spec, apply changes to main specs**
+
+   For each repo-local capability delta spec path returned by the CLI:
+
+   a. **Read the delta spec** to understand the intended changes
+
+   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
+
+   c. **Apply changes intelligently**:
+
+      **ADDED Requirements:**
+      - If requirement doesn't exist in main spec → add it
+      - If requirement already exists → update it to match (treat as implicit MODIFIED)
+
+      **MODIFIED Requirements:**
+      - Find the requirement in main spec
+      - Apply the changes - this can be:
+        - Adding new scenarios (don't need to copy existing ones)
+        - Modifying existing scenarios
+        - Changing the requirement description
+      - Preserve scenarios/content not mentioned in the delta
+
+      **REMOVED Requirements:**
+      - Remove the entire requirement block from main spec
+
+      **RENAMED Requirements:**
+      - Find the FROM requirement, rename to TO
+
+   d. **Create new main spec** if capability doesn't exist yet:
+      - Create `openspec/specs/<capability>/spec.md`
+      - Add Purpose section (can be brief, mark as TBD)
+      - Add Requirements section with the ADDED requirements
+
+5. **Show summary**
+
+   After applying all changes, summarize:
+   - Which capabilities were updated
+   - What changes were made (requirements added/modified/removed/renamed)
+
+**Delta Spec Format Reference**
+
+```markdown
+## ADDED Requirements
+
+### Requirement: New Feature
+The system SHALL do something new.
+
+#### Scenario: Basic case
+- **WHEN** user does X
+- **THEN** system does Y
+
+## MODIFIED Requirements
+
+### Requirement: Existing Feature
+#### Scenario: New scenario to add
+- **WHEN** user does A
+- **THEN** system does B
+
+## REMOVED Requirements
+
+### Requirement: Deprecated Feature
+
+## RENAMED Requirements
+
+- FROM: `### Requirement: Old Name`
+- TO: `### Requirement: New Name`
+```
+
+**Key Principle: Intelligent Merging**
+
+Unlike programmatic merging, you can apply **partial updates**:
+- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
+- The delta represents *intent*, not a wholesale replacement
+- Use your judgment to merge changes sensibly
+
+**Output On Success**
+
+```
+## Specs Synced: <change-name>
+
+Updated main specs:
+
+**<capability-1>**:
+- Added requirement: "New Feature"
+- Modified requirement: "Existing Feature" (added 1 scenario)
+
+**<capability-2>**:
+- Created new spec file
+- Added requirement: "Another Feature"
+
+Main specs are now updated. The change remains active - archive when implementation is complete.
+```
+
+**Guardrails**
+- Read both delta and main specs before making changes
+- Preserve existing content not mentioned in delta
+- If something is unclear, ask for clarification
+- Show what you're changing as you go
+- The operation should be idempotent - running twice should give same result
diff --git a/packages/.github/skills/openspec-update-change/SKILL.md b/packages/.github/skills/openspec-update-change/SKILL.md
new file mode 100644
index 00000000..3e1c931d
--- /dev/null
+++ b/packages/.github/skills/openspec-update-change/SKILL.md
@@ -0,0 +1,86 @@
+---
+name: openspec-update-change
+description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Revise a change's existing planning artifacts and keep them coherent. Never edit code.
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
+
+   Present the top 3-4 most recently modified changes as options, showing:
+   - Change name
+   - Schema (from `schema` field if present, otherwise "spec-driven")
+   - Status (e.g., "0/5 tasks", "complete", "no tasks")
+   - How recently it was modified (from `lastModified` field)
+
+   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Get the change's artifacts**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand current state. The response includes:
+   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
+   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
+   - `isComplete`: Boolean indicating if all artifacts are complete
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
+
+   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
+
+   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
+
+3. **Understand the request**
+   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
+   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
+
+4. **Read and reconcile**
+   - Read the artifact(s) the request touches and the change's other existing artifacts.
+   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
+   - Note everything that is now inconsistent, missing, or contradictory.
+   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
+   - If the change is already coherent, say so and make no edits.
+
+5. **Confirm and apply, one artifact at a time**
+   - Show each proposed revision and why. Write only after the user confirms.
+   - If the user rejects a revision, do not write it - leave that artifact unchanged.
+   - When a substantial rewrite is needed, get that artifact's rules and template first:
+     ```bash
+     openspec instructions <artifact-id> --change "<name>" --json
+     ```
+
+6. **Point to the next step (guidance only - NEVER act on it)**
+   - Artifacts still missing -> suggest `/opsx:continue` to create them.
+   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
+   - Everything done and implemented -> suggest `/opsx:archive`.
+
+**Output**
+
+After each invocation, show:
+- Which artifacts were revised (and which proposed revisions were rejected)
+- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
+- Where the change stands and the recommended next command
+
+**Guardrails**
+- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
+- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
+- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
+- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
+- Confirm every edit with the user before writing.
+- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.github/skills/openspec-verify-change/SKILL.md b/packages/.github/skills/openspec-verify-change/SKILL.md
new file mode 100644
index 00000000..b6557c6f
--- /dev/null
+++ b/packages/.github/skills/openspec-verify-change/SKILL.md
@@ -0,0 +1,172 @@
+---
+name: openspec-verify-change
+description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
+allowed-tools: Bash(openspec:*)
+license: MIT
+compatibility: Requires openspec CLI.
+metadata:
+  author: openspec
+  version: "1.0"
+  generatedBy: "1.6.0"
+---
+
+Verify that an implementation matches the change artifacts (specs, tasks, design).
+
+**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
+
+**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
+
+**Steps**
+
+1. **If no change name provided, prompt for selection**
+
+   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
+
+   Show changes that have implementation tasks (tasks artifact exists).
+   Include the schema used for each change if available.
+   Mark changes with incomplete tasks as "(In Progress)".
+
+   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
+
+2. **Check status to understand the schema**
+   ```bash
+   openspec status --change "<name>" --json
+   ```
+   Parse the JSON to understand:
+   - `schemaName`: The workflow being used (e.g., "spec-driven")
+   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
+   - Which artifacts exist for this change
+
+3. **Get planning context and load artifacts**
+
+   ```bash
+   openspec instructions apply --change "<name>" --json
+   ```
+
+   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
+
+4. **Initialize verification report structure**
+
+   Create a report structure with three dimensions:
+   - **Completeness**: Track tasks and spec coverage
+   - **Correctness**: Track requirement implementation and scenario coverage
+   - **Coherence**: Track design adherence and pattern consistency
+
+   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
+
+5. **Verify Completeness**
+
+   **Task Completion**:
+   - If `contextFiles.tasks` exists, read every file path in it
+   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
+   - Count complete vs total tasks
+   - If incomplete tasks exist:
+     - Add CRITICAL issue for each incomplete task
+     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
+
+   **Spec Coverage**:
+   - If delta specs exist in `contextFiles.specs`:
+     - Extract all requirements (marked with "### Requirement:")
+     - For each requirement:
+       - Search codebase for keywords related to the requirement
+       - Assess if implementation likely exists
+     - If requirements appear unimplemented:
+       - Add CRITICAL issue: "Requirement not found: <requirement name>"
+       - Recommendation: "Implement requirement X: <description>"
+
+6. **Verify Correctness**
+
+   **Requirement Implementation Mapping**:
+   - For each requirement from delta specs:
+     - Search codebase for implementation evidence
+     - If found, note file paths and line ranges
+     - Assess if implementation matches requirement intent
+     - If divergence detected:
+       - Add WARNING: "Implementation may diverge from spec: <details>"
+       - Recommendation: "Review <file>:<lines> against requirement X"
+
+   **Scenario Coverage**:
+   - For each scenario in delta specs (marked with "#### Scenario:"):
+     - Check if conditions are handled in code
+     - Check if tests exist covering the scenario
+     - If scenario appears uncovered:
+       - Add WARNING: "Scenario not covered: <scenario name>"
+       - Recommendation: "Add test or implementation for scenario: <description>"
+
+7. **Verify Coherence**
+
+   **Design Adherence**:
+   - If `contextFiles.design` exists:
+     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
+     - Verify implementation follows those decisions
+     - If contradiction detected:
+       - Add WARNING: "Design decision not followed: <decision>"
+       - Recommendation: "Update implementation or revise design.md to match reality"
+   - If no design.md: Skip design adherence check, note "No design.md to verify against"
+
+   **Code Pattern Consistency**:
+   - Review new code for consistency with project patterns
+   - Check file naming, directory structure, coding style
+   - If significant deviations found:
+     - Add SUGGESTION: "Code pattern deviation: <details>"
+     - Recommendation: "Consider following project pattern: <example>"
+
+8. **Generate Verification Report**
+
+   **Summary Scorecard**:
+   ```
+   ## Verification Report: <change-name>
+
+   ### Summary
+   | Dimension    | Status           |
+   |--------------|------------------|
+   | Completeness | X/Y tasks, N reqs|
+   | Correctness  | M/N reqs covered |
+   | Coherence    | Followed/Issues  |
+   ```
+
+   **Issues by Priority**:
+
+   1. **CRITICAL** (Must fix before archive):
+      - Incomplete tasks
+      - Missing requirement implementations
+      - Each with specific, actionable recommendation
+
+   2. **WARNING** (Should fix):
+      - Spec/design divergences
+      - Missing scenario coverage
+      - Each with specific recommendation
+
+   3. **SUGGESTION** (Nice to fix):
+      - Pattern inconsistencies
+      - Minor improvements
+      - Each with specific recommendation
+
+   **Final Assessment**:
+   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
+   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
+   - If all clear: "All checks passed. Ready for archive."
+
+**Verification Heuristics**
+
+- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
+- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
+- **Coherence**: Look for glaring inconsistencies, don't nitpick style
+- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
+- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
+
+**Graceful Degradation**
+
+- If only tasks.md exists: verify task completion only, skip spec/design checks
+- If tasks + specs exist: verify completeness and correctness, skip design
+- If full artifacts: verify all three dimensions
+- Always note which checks were skipped and why
+
+**Output Format**
+
+Use clear markdown with:
+- Table for summary scorecard
+- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
+- Code references in format: `file.ts:123`
+- Specific, actionable recommendations
+- No vague suggestions like "consider reviewing"
diff --git a/packages/openspec/config.yaml b/packages/openspec/config.yaml
new file mode 100644
index 00000000..392946c6
--- /dev/null
+++ b/packages/openspec/config.yaml
@@ -0,0 +1,20 @@
+schema: spec-driven
+
+# Project context (optional)
+# This is shown to AI when creating artifacts.
+# Add your tech stack, conventions, style guides, domain knowledge, etc.
+# Example:
+#   context: |
+#     Tech stack: TypeScript, React, Node.js
+#     We use conventional commits
+#     Domain: e-commerce platform
+
+# Per-artifact rules (optional)
+# Add custom rules for specific artifacts.
+# Example:
+#   rules:
+#     proposal:
+#       - Keep proposals under 500 words
+#       - Always include a "Non-goals" section
+#     tasks:
+#       - Break tasks into chunks of max 2 hours

From 0be9f41c323288edab7e13a6842a97eb0e2522c7 Mon Sep 17 00:00:00 2001
From: George Berezhnoy <gohabereg@users.noreply.github.com>
Date: Sat, 18 Jul 2026 00:15:06 +0100
Subject: [PATCH 2/5] Update openspec/config.yaml

Co-authored-by: Reversean <43142829+Reversean@users.noreply.github.com>
---
 openspec/config.yaml | 37 ++++++++++++++++++++-----------------
 1 file changed, 20 insertions(+), 17 deletions(-)

diff --git a/openspec/config.yaml b/openspec/config.yaml
index 392946c6..6de0e731 100644
--- a/openspec/config.yaml
+++ b/openspec/config.yaml
@@ -1,20 +1,23 @@
 schema: spec-driven
 
-# Project context (optional)
-# This is shown to AI when creating artifacts.
-# Add your tech stack, conventions, style guides, domain knowledge, etc.
-# Example:
-#   context: |
-#     Tech stack: TypeScript, React, Node.js
-#     We use conventional commits
-#     Domain: e-commerce platform
+context: |
+  Tech stack: TypeScript, Yarn 4 (Berry) workspaces monorepo (packages/*, packages/tools/*).
+  Domain: Editor.js document model - a collaborative rich-text editor engine (block/inline
+  data tree, operational-transform based collaboration, pluggable block/inline tools).
+  Key packages: model (EditorJSModel/EditorDocument/BlockNode/inline-fragments/CaretManager),
+  model-types (shared event/type contracts), core, dom-adapters, ot-server (collaboration/OT
+  server), sdk (public API for building tools/plugins - tools should depend on this, not on
+  model directly), ui, collaboration-manager, and packages/tools/* (built-in block/inline
+  tools implementing BlockTool/InlineTool from @editorjs/sdk).
+  Existing architecture docs live in docs/; Mermaid diagrams in docs/diagrams/;
+  consult these before writing new specs/proposals so they don't duplicate or contradict
+  what's already documented there.
+  Testing: Jest with co-located *.spec.ts/*.integration.spec.ts files; some packages also
+  run Stryker mutation testing (test:mutations). Linting via ESLint (yarn lint).
 
-# Per-artifact rules (optional)
-# Add custom rules for specific artifacts.
-# Example:
-#   rules:
-#     proposal:
-#       - Keep proposals under 500 words
-#       - Always include a "Non-goals" section
-#     tasks:
-#       - Break tasks into chunks of max 2 hours
+rules:
+  tasks:
+    - Write the failing test first, then implement, then refactor (TDD).
+    - Name test cases in should-notation, e.g. it('should return X when Y').
+  proposal:
+    - Cross-check against existing docs in docs/ and note if the proposal supersedes any of them.

From ec47929f98f6f76f364cc5d3f16405a88139a931 Mon Sep 17 00:00:00 2001
From: gohabereg <gohabereg@gmail.com>
Date: Sat, 18 Jul 2026 00:16:48 +0100
Subject: [PATCH 3/5] Remove duplicated skills and prompts

---
 .github/prompts/opsx-apply.prompt.md          | 152 -----
 .github/prompts/opsx-archive.prompt.md        | 157 -----
 .github/prompts/opsx-explore.prompt.md        | 171 ------
 .github/prompts/opsx-onboard.prompt.md        | 547 -----------------
 .github/prompts/opsx-propose.prompt.md        | 106 ----
 .github/prompts/opsx-sync.prompt.md           | 140 -----
 .github/prompts/opsx-update.prompt.md         |  78 ---
 .github/prompts/opsx-verify.prompt.md         | 164 ------
 .github/skills/openspec-apply-change/SKILL.md | 160 -----
 .../skills/openspec-archive-change/SKILL.md   | 118 ----
 .github/skills/openspec-explore/SKILL.md      | 290 ---------
 .github/skills/openspec-onboard/SKILL.md      | 555 ------------------
 .github/skills/openspec-propose/SKILL.md      | 114 ----
 .github/skills/openspec-sync-specs/SKILL.md   | 148 -----
 .../skills/openspec-update-change/SKILL.md    |  86 ---
 .../skills/openspec-verify-change/SKILL.md    | 172 ------
 .opencode/commands/opsx-apply.md              | 152 -----
 .opencode/commands/opsx-archive.md            | 157 -----
 .opencode/commands/opsx-explore.md            | 171 ------
 .opencode/commands/opsx-onboard.md            | 547 -----------------
 .opencode/commands/opsx-propose.md            | 106 ----
 .opencode/commands/opsx-sync.md               | 140 -----
 .opencode/commands/opsx-update.md             |  78 ---
 .opencode/commands/opsx-verify.md             | 164 ------
 .../skills/openspec-apply-change/SKILL.md     | 160 -----
 .../skills/openspec-archive-change/SKILL.md   | 118 ----
 .opencode/skills/openspec-explore/SKILL.md    | 290 ---------
 .opencode/skills/openspec-onboard/SKILL.md    | 555 ------------------
 .opencode/skills/openspec-propose/SKILL.md    | 114 ----
 .opencode/skills/openspec-sync-specs/SKILL.md | 148 -----
 .../skills/openspec-update-change/SKILL.md    |  86 ---
 .../skills/openspec-verify-change/SKILL.md    | 172 ------
 32 files changed, 6316 deletions(-)
 delete mode 100644 .github/prompts/opsx-apply.prompt.md
 delete mode 100644 .github/prompts/opsx-archive.prompt.md
 delete mode 100644 .github/prompts/opsx-explore.prompt.md
 delete mode 100644 .github/prompts/opsx-onboard.prompt.md
 delete mode 100644 .github/prompts/opsx-propose.prompt.md
 delete mode 100644 .github/prompts/opsx-sync.prompt.md
 delete mode 100644 .github/prompts/opsx-update.prompt.md
 delete mode 100644 .github/prompts/opsx-verify.prompt.md
 delete mode 100644 .github/skills/openspec-apply-change/SKILL.md
 delete mode 100644 .github/skills/openspec-archive-change/SKILL.md
 delete mode 100644 .github/skills/openspec-explore/SKILL.md
 delete mode 100644 .github/skills/openspec-onboard/SKILL.md
 delete mode 100644 .github/skills/openspec-propose/SKILL.md
 delete mode 100644 .github/skills/openspec-sync-specs/SKILL.md
 delete mode 100644 .github/skills/openspec-update-change/SKILL.md
 delete mode 100644 .github/skills/openspec-verify-change/SKILL.md
 delete mode 100644 .opencode/commands/opsx-apply.md
 delete mode 100644 .opencode/commands/opsx-archive.md
 delete mode 100644 .opencode/commands/opsx-explore.md
 delete mode 100644 .opencode/commands/opsx-onboard.md
 delete mode 100644 .opencode/commands/opsx-propose.md
 delete mode 100644 .opencode/commands/opsx-sync.md
 delete mode 100644 .opencode/commands/opsx-update.md
 delete mode 100644 .opencode/commands/opsx-verify.md
 delete mode 100644 .opencode/skills/openspec-apply-change/SKILL.md
 delete mode 100644 .opencode/skills/openspec-archive-change/SKILL.md
 delete mode 100644 .opencode/skills/openspec-explore/SKILL.md
 delete mode 100644 .opencode/skills/openspec-onboard/SKILL.md
 delete mode 100644 .opencode/skills/openspec-propose/SKILL.md
 delete mode 100644 .opencode/skills/openspec-sync-specs/SKILL.md
 delete mode 100644 .opencode/skills/openspec-update-change/SKILL.md
 delete mode 100644 .opencode/skills/openspec-verify-change/SKILL.md

diff --git a/.github/prompts/opsx-apply.prompt.md b/.github/prompts/opsx-apply.prompt.md
deleted file mode 100644
index 5bd0b6bb..00000000
--- a/.github/prompts/opsx-apply.prompt.md
+++ /dev/null
@@ -1,152 +0,0 @@
----
-description: Implement tasks from an OpenSpec change (Experimental)
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! You can archive this change with `/opsx:archive`.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.github/prompts/opsx-archive.prompt.md b/.github/prompts/opsx-archive.prompt.md
deleted file mode 100644
index f9ba89d8..00000000
--- a/.github/prompts/opsx-archive.prompt.md
+++ /dev/null
@@ -1,157 +0,0 @@
----
-description: Archive a completed change in the experimental workflow
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Spec sync status (synced / sync skipped / no delta specs)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success (No Delta Specs)**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** No delta specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success With Warnings**
-
-```
-## Archive Complete (with warnings)
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** Sync skipped (user chose to skip)
-
-**Warnings:**
-- Archived with 2 incomplete artifacts
-- Archived with 3 incomplete tasks
-- Delta spec sync was skipped (user chose to skip)
-
-Review the archive if this was not intentional.
-```
-
-**Output On Error (Archive Exists)**
-
-```
-## Archive Failed
-
-**Change:** <change-name>
-**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-
-Target archive directory already exists.
-
-**Options:**
-1. Rename the existing archive
-2. Delete the existing archive if it's a duplicate
-3. Wait until a different date to archive
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.github/prompts/opsx-explore.prompt.md b/.github/prompts/opsx-explore.prompt.md
deleted file mode 100644
index 856001a6..00000000
--- a/.github/prompts/opsx-explore.prompt.md
+++ /dev/null
@@ -1,171 +0,0 @@
----
-description: Enter explore mode - think through ideas, investigate problems, clarify requirements
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
-- A vague idea: "real-time collaboration"
-- A specific problem: "the auth system is getting unwieldy"
-- A change name: "add-dark-mode" (to explore in context of that change)
-- A comparison: "postgres vs sqlite for this"
-- Nothing (just enter explore mode)
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-If the user mentioned a specific change name, read its artifacts for context.
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/.github/prompts/opsx-onboard.prompt.md b/.github/prompts/opsx-onboard.prompt.md
deleted file mode 100644
index 84aa48d6..00000000
--- a/.github/prompts/opsx-onboard.prompt.md
+++ /dev/null
@@ -1,547 +0,0 @@
----
-description: Guided onboarding - walk through a complete OpenSpec workflow cycle with narration
----
-
-Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## Preflight
-
-Before starting, check if the OpenSpec CLI is installed:
-
-```bash
-# Unix/macOS
-openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
-# Windows (PowerShell)
-# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
-```
-
-**If CLI not installed:**
-> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
-
-Stop here if not installed.
-
----
-
-## Phase 1: Welcome
-
-Display:
-
-```
-## Welcome to OpenSpec!
-
-I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
-
-**What we'll do:**
-1. Pick a small, real task in your codebase
-2. Explore the problem briefly
-3. Create a change (the container for our work)
-4. Build the artifacts: proposal → specs → design → tasks
-5. Implement the tasks
-6. Archive the completed change
-
-**Time:** ~15-20 minutes
-
-Let's start by finding something to work on.
-```
-
----
-
-## Phase 2: Task Selection
-
-### Codebase Analysis
-
-Scan the codebase for small improvement opportunities. Look for:
-
-1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
-2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
-3. **Functions without tests** - Cross-reference `src/` with test directories
-4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
-5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
-6. **Missing validation** - User input handlers without validation
-
-Also check recent git activity:
-```bash
-# Unix/macOS
-git log --oneline -10 2>/dev/null || echo "No git history"
-# Windows (PowerShell)
-# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
-```
-
-### Present Suggestions
-
-From your analysis, present 3-4 specific suggestions:
-
-```
-## Task Suggestions
-
-Based on scanning your codebase, here are some good starter tasks:
-
-**1. [Most promising task]**
-   Location: `src/path/to/file.ts:42`
-   Scope: ~1-2 files, ~20-30 lines
-   Why it's good: [brief reason]
-
-**2. [Second task]**
-   Location: `src/another/file.ts`
-   Scope: ~1 file, ~15 lines
-   Why it's good: [brief reason]
-
-**3. [Third task]**
-   Location: [location]
-   Scope: [estimate]
-   Why it's good: [brief reason]
-
-**4. Something else?**
-   Tell me what you'd like to work on.
-
-Which task interests you? (Pick a number or describe your own)
-```
-
-**If nothing found:** Fall back to asking what the user wants to build:
-> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
-
-### Scope Guardrail
-
-If the user picks or describes something too large (major feature, multi-day work):
-
-```
-That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
-
-For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
-
-**Options:**
-1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
-2. **Pick something else** - One of the other suggestions, or a different small task?
-3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
-
-What would you prefer?
-```
-
-Let the user override if they insist—this is a soft guardrail.
-
----
-
-## Phase 3: Explore Demo
-
-Once a task is selected, briefly demonstrate explore mode:
-
-```
-Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
-```
-
-Spend 1-2 minutes investigating the relevant code:
-- Read the file(s) involved
-- Draw a quick ASCII diagram if it helps
-- Note any considerations
-
-```
-## Quick Exploration
-
-[Your brief analysis—what you found, any considerations]
-
-┌─────────────────────────────────────────┐
-│   [Optional: ASCII diagram if helpful]  │
-└─────────────────────────────────────────┘
-
-Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
-
-Now let's create a change to hold our work.
-```
-
-**PAUSE** - Wait for user acknowledgment before proceeding.
-
----
-
-## Phase 4: Create the Change
-
-**EXPLAIN:**
-```
-## Creating a Change
-
-A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
-
-Let me create one for our task.
-```
-
-**DO:** Create the change with a derived kebab-case name:
-```bash
-openspec new change "<derived-name>"
-```
-
-**SHOW:**
-```
-Created: <changeRoot from status JSON>
-
-The folder structure:
-```
-<changeRoot>/
-├── proposal.md    ← Why we're doing this (empty, we'll fill it)
-├── design.md      ← How we'll build it (empty)
-├── specs/         ← Detailed requirements (empty)
-└── tasks.md       ← Implementation checklist (empty)
-```
-
-Now let's fill in the first artifact—the proposal.
-```
-
----
-
-## Phase 5: Proposal
-
-**EXPLAIN:**
-```
-## The Proposal
-
-The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
-
-I'll draft one based on our task.
-```
-
-**DO:** Draft the proposal content (don't save yet):
-
-```
-Here's a draft proposal:
-
----
-
-## Why
-
-[1-2 sentences explaining the problem/opportunity]
-
-## What Changes
-
-[Bullet points of what will be different]
-
-## Capabilities
-
-### New Capabilities
-- `<capability-name>`: [brief description]
-
-### Modified Capabilities
-<!-- If modifying existing behavior -->
-
-## Impact
-
-- `src/path/to/file.ts`: [what changes]
-- [other files if applicable]
-
----
-
-Does this capture the intent? I can adjust before we save it.
-```
-
-**PAUSE** - Wait for user approval/feedback.
-
-After approval, save the proposal:
-```bash
-openspec instructions proposal --change "<name>" --json
-```
-Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
-
-```
-Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
-
-Next up: specs.
-```
-
----
-
-## Phase 6: Specs
-
-**EXPLAIN:**
-```
-## Specs
-
-Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
-
-For a small task like this, we might only need one spec file.
-```
-
-**DO:** Resolve where the spec file should be created:
-```bash
-openspec instructions specs --change "<name>" --json
-# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
-```
-
-Draft the spec content:
-
-```
-Here's the spec:
-
----
-
-## ADDED Requirements
-
-### Requirement: <Name>
-
-<Description of what the system should do>
-
-#### Scenario: <Scenario name>
-
-- **WHEN** <trigger condition>
-- **THEN** <expected outcome>
-- **AND** <additional outcome if needed>
-
----
-
-This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
-```
-
-Save to the concrete file path chosen from `resolvedOutputPath`.
-
----
-
-## Phase 7: Design
-
-**EXPLAIN:**
-```
-## Design
-
-The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
-
-For small changes, this might be brief. That's fine—not every change needs deep design discussion.
-```
-
-**DO:** Draft design.md:
-
-```
-Here's the design:
-
----
-
-## Context
-
-[Brief context about the current state]
-
-## Goals / Non-Goals
-
-**Goals:**
-- [What we're trying to achieve]
-
-**Non-Goals:**
-- [What's explicitly out of scope]
-
-## Decisions
-
-### Decision 1: [Key decision]
-
-[Explanation of approach and rationale]
-
----
-
-For a small task, this captures the key decisions without over-engineering.
-```
-
-Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
-
----
-
-## Phase 8: Tasks
-
-**EXPLAIN:**
-```
-## Tasks
-
-Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
-
-These should be small, clear, and in logical order.
-```
-
-**DO:** Generate tasks based on specs and design:
-
-```
-Here are the implementation tasks:
-
----
-
-## 1. [Category or file]
-
-- [ ] 1.1 [Specific task]
-- [ ] 1.2 [Specific task]
-
-## 2. Verify
-
-- [ ] 2.1 [Verification step]
-
----
-
-Each checkbox becomes a unit of work in the apply phase. Ready to implement?
-```
-
-**PAUSE** - Wait for user to confirm they're ready to implement.
-
-Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
-
----
-
-## Phase 9: Apply (Implementation)
-
-**EXPLAIN:**
-```
-## Implementation
-
-Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
-```
-
-**DO:** For each task:
-
-1. Announce: "Working on task N: [description]"
-2. Implement the change in the codebase
-3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
-4. Mark complete in tasks.md: `- [ ]` → `- [x]`
-5. Brief status: "✓ Task N complete"
-
-Keep narration light—don't over-explain every line of code.
-
-After all tasks:
-
-```
-## Implementation Complete
-
-All tasks done:
-- [x] Task 1
-- [x] Task 2
-- [x] ...
-
-The change is implemented! One more step—let's archive it.
-```
-
----
-
-## Phase 10: Archive
-
-**EXPLAIN:**
-```
-## Archiving
-
-When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
-
-Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
-```
-
-**DO:**
-```bash
-openspec archive "<name>"
-```
-
-**SHOW:**
-```
-Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
-
-The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
-```
-
----
-
-## Phase 11: Recap & Next Steps
-
-```
-## Congratulations!
-
-You just completed a full OpenSpec cycle:
-
-1. **Explore** - Thought through the problem
-2. **New** - Created a change container
-3. **Proposal** - Captured WHY
-4. **Specs** - Defined WHAT in detail
-5. **Design** - Decided HOW
-6. **Tasks** - Broke it into steps
-7. **Apply** - Implemented the work
-8. **Archive** - Preserved the record
-
-This same rhythm works for any size change—a small fix or a major feature.
-
----
-
-## Command Reference
-
-**Core workflow:**
-
- | Command           | What it does                               |
- |-------------------|--------------------------------------------|
- | `/opsx:propose` | Create a change and generate all artifacts |
- | `/opsx:explore` | Think through problems before/during work  |
- | `/opsx:apply`   | Implement tasks from a change              |
- | `/opsx:archive` | Archive a completed change                 |
-
-**Additional commands:**
-
- | Command            | What it does                                             |
- |--------------------|----------------------------------------------------------|
- | `/opsx:new`      | Start a new change, step through artifacts one at a time |
- | `/opsx:continue` | Continue working on an existing change                   |
- | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
- | `/opsx:verify`   | Verify implementation matches artifacts                  |
-
----
-
-## What's Next?
-
-Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
-```
-
----
-
-## Graceful Exit Handling
-
-### User wants to stop mid-way
-
-If the user says they need to stop, want to pause, or seem disengaged:
-
-```
-No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
-
-To pick up where we left off later:
-- `/opsx:continue <name>` - Resume artifact creation
-- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
-
-The work won't be lost. Come back whenever you're ready.
-```
-
-Exit gracefully without pressure.
-
-### User just wants command reference
-
-If the user says they just want to see the commands or skip the tutorial:
-
-```
-## OpenSpec Quick Reference
-
-**Core workflow:**
-
- | Command                  | What it does                               |
- |--------------------------|--------------------------------------------|
- | `/opsx:propose <name>` | Create a change and generate all artifacts |
- | `/opsx:explore`        | Think through problems (no code changes)   |
- | `/opsx:apply <name>`   | Implement tasks                            |
- | `/opsx:archive <name>` | Archive when done                          |
-
-**Additional commands:**
-
- | Command                   | What it does                        |
- |---------------------------|-------------------------------------|
- | `/opsx:new <name>`      | Start a new change, step by step    |
- | `/opsx:continue <name>` | Continue an existing change         |
- | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
- | `/opsx:verify <name>`   | Verify implementation               |
-
-Try `/opsx:propose` to start your first change.
-```
-
-Exit gracefully.
-
----
-
-## Guardrails
-
-- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
-- **Keep narration light** during implementation—teach without lecturing
-- **Don't skip phases** even if the change is small—the goal is teaching the workflow
-- **Pause for acknowledgment** at marked points, but don't over-pause
-- **Handle exits gracefully**—never pressure the user to continue
-- **Use real codebase tasks**—don't simulate or use fake examples
-- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.github/prompts/opsx-propose.prompt.md b/.github/prompts/opsx-propose.prompt.md
deleted file mode 100644
index fde349a6..00000000
--- a/.github/prompts/opsx-propose.prompt.md
+++ /dev/null
@@ -1,106 +0,0 @@
----
-description: Propose a new change - create it and generate all artifacts in one step
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
-
-**Steps**
-
-1. **If no input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` to start implementing."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/.github/prompts/opsx-sync.prompt.md b/.github/prompts/opsx-sync.prompt.md
deleted file mode 100644
index 961aa409..00000000
--- a/.github/prompts/opsx-sync.prompt.md
+++ /dev/null
@@ -1,140 +0,0 @@
----
-description: Sync delta specs from a change to main specs
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/.github/prompts/opsx-update.prompt.md b/.github/prompts/opsx-update.prompt.md
deleted file mode 100644
index 5b316d4d..00000000
--- a/.github/prompts/opsx-update.prompt.md
+++ /dev/null
@@ -1,78 +0,0 @@
----
-description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.github/prompts/opsx-verify.prompt.md b/.github/prompts/opsx-verify.prompt.md
deleted file mode 100644
index cc1377a7..00000000
--- a/.github/prompts/opsx-verify.prompt.md
+++ /dev/null
@@ -1,164 +0,0 @@
----
-description: Verify implementation matches change artifacts before archiving
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/.github/skills/openspec-apply-change/SKILL.md b/.github/skills/openspec-apply-change/SKILL.md
deleted file mode 100644
index 1375861c..00000000
--- a/.github/skills/openspec-apply-change/SKILL.md
+++ /dev/null
@@ -1,160 +0,0 @@
----
-name: openspec-apply-change
-description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! Ready to archive this change.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.github/skills/openspec-archive-change/SKILL.md b/.github/skills/openspec-archive-change/SKILL.md
deleted file mode 100644
index c0c169d1..00000000
--- a/.github/skills/openspec-archive-change/SKILL.md
+++ /dev/null
@@ -1,118 +0,0 @@
----
-name: openspec-archive-change
-description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Whether specs were synced (if applicable)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
-
-All artifacts complete. All tasks complete.
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use openspec-sync-specs approach (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.github/skills/openspec-explore/SKILL.md b/.github/skills/openspec-explore/SKILL.md
deleted file mode 100644
index 771271a8..00000000
--- a/.github/skills/openspec-explore/SKILL.md
+++ /dev/null
@@ -1,290 +0,0 @@
----
-name: openspec-explore
-description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Handling Different Entry Points
-
-**User brings a vague idea:**
-```
-User: I'm thinking about adding real-time collaboration
-
-You: Real-time collab is a big space. Let me think about this...
-
-      COLLABORATION SPECTRUM
-      ════════════════════════════════════════════
-
-      Awareness          Coordination         Sync
-          │                   │                 │
-          ▼                   ▼                 ▼
-      ┌────────┐         ┌────────┐        ┌────────┐
-      │Presence│         │Cursors │        │  CRDT  │
-      │  "3    │         │ Multi  │        │Conflict│
-      │online" │         │ select │        │  free  │
-      └────────┘         └────────┘        └────────┘
-          │                   │                 │
-       trivial            moderate           complex
-
-      Where's your head at?
-```
-
-**User brings a specific problem:**
-```
-User: The auth system is a mess
-
-You: [reads codebase]
-
-     ┌─────────────────────────────────────────────┐
-     │              CURRENT AUTH FLOW              │
-     └─────────────────────────────────────────────┘
-                        │
-          ┌─────────────┼─────────────┐
-          ▼             ▼             ▼
-     ┌─────────┐  ┌─────────┐  ┌─────────┐
-     │ Google  │  │  GitHub │  │  Email  │
-     │  OAuth  │  │  OAuth  │  │  Magic  │
-     └────┬────┘  └────┬────┘  └────┬────┘
-          │             │             │
-          └─────────────┼─────────────┘
-                        ▼
-                  ┌───────────┐
-                  │  Session  │
-                  └─────┬─────┘
-                        │
-                        ▼
-                  ┌───────────┐
-                  │   Perms   │
-                  └───────────┘
-
-     I see three tangles. Which one's burning?
-```
-
-**User is stuck mid-implementation:**
-```
-User: /opsx:explore add-auth-system
-      The OAuth integration is more complex than expected
-
-You: [reads change artifacts]
-
-     You're on task 4: "Implement OAuth flow"
-
-     Let me trace what's involved...
-
-     [draws diagram, explores options, suggests paths]
-
-     Want to update the design to reflect this?
-     Or add a spike task to investigate?
-```
-
-**User wants to compare options:**
-```
-User: Should we use Postgres or SQLite?
-
-You: Generic answer is boring. What's the context?
-
-User: A CLI tool that tracks local dev environments
-
-You: That changes everything.
-
-     ┌─────────────────────────────────────────────────┐
-     │          CLI TOOL DATA STORAGE                  │
-     └─────────────────────────────────────────────────┘
-
-     Key constraints:
-     • No daemon running
-     • Must work offline
-     • Single user
-
-                  SQLite          Postgres
-     Deployment   embedded ✓      needs server ✗
-     Offline      yes ✓           no ✗
-     Single file  yes ✓           no ✗
-
-     SQLite. Not even close.
-
-     Unless... is there a sync component?
-```
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When it feels like things are crystallizing, you might summarize:
-
-```
-## What We Figured Out
-
-**The problem**: [crystallized understanding]
-
-**The approach**: [if one emerged]
-
-**Open questions**: [if any remain]
-
-**Next steps** (if ready):
-- Create a change proposal
-- Keep exploring: just keep talking
-```
-
-But this summary is optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/.github/skills/openspec-onboard/SKILL.md b/.github/skills/openspec-onboard/SKILL.md
deleted file mode 100644
index ac4ac6fe..00000000
--- a/.github/skills/openspec-onboard/SKILL.md
+++ /dev/null
@@ -1,555 +0,0 @@
----
-name: openspec-onboard
-description: Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## Preflight
-
-Before starting, check if the OpenSpec CLI is installed:
-
-```bash
-# Unix/macOS
-openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
-# Windows (PowerShell)
-# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
-```
-
-**If CLI not installed:**
-> OpenSpec CLI is not installed. Install it first, then come back to `/opsx:onboard`.
-
-Stop here if not installed.
-
----
-
-## Phase 1: Welcome
-
-Display:
-
-```
-## Welcome to OpenSpec!
-
-I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
-
-**What we'll do:**
-1. Pick a small, real task in your codebase
-2. Explore the problem briefly
-3. Create a change (the container for our work)
-4. Build the artifacts: proposal → specs → design → tasks
-5. Implement the tasks
-6. Archive the completed change
-
-**Time:** ~15-20 minutes
-
-Let's start by finding something to work on.
-```
-
----
-
-## Phase 2: Task Selection
-
-### Codebase Analysis
-
-Scan the codebase for small improvement opportunities. Look for:
-
-1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
-2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
-3. **Functions without tests** - Cross-reference `src/` with test directories
-4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
-5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
-6. **Missing validation** - User input handlers without validation
-
-Also check recent git activity:
-```bash
-# Unix/macOS
-git log --oneline -10 2>/dev/null || echo "No git history"
-# Windows (PowerShell)
-# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
-```
-
-### Present Suggestions
-
-From your analysis, present 3-4 specific suggestions:
-
-```
-## Task Suggestions
-
-Based on scanning your codebase, here are some good starter tasks:
-
-**1. [Most promising task]**
-   Location: `src/path/to/file.ts:42`
-   Scope: ~1-2 files, ~20-30 lines
-   Why it's good: [brief reason]
-
-**2. [Second task]**
-   Location: `src/another/file.ts`
-   Scope: ~1 file, ~15 lines
-   Why it's good: [brief reason]
-
-**3. [Third task]**
-   Location: [location]
-   Scope: [estimate]
-   Why it's good: [brief reason]
-
-**4. Something else?**
-   Tell me what you'd like to work on.
-
-Which task interests you? (Pick a number or describe your own)
-```
-
-**If nothing found:** Fall back to asking what the user wants to build:
-> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
-
-### Scope Guardrail
-
-If the user picks or describes something too large (major feature, multi-day work):
-
-```
-That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
-
-For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
-
-**Options:**
-1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
-2. **Pick something else** - One of the other suggestions, or a different small task?
-3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
-
-What would you prefer?
-```
-
-Let the user override if they insist—this is a soft guardrail.
-
----
-
-## Phase 3: Explore Demo
-
-Once a task is selected, briefly demonstrate explore mode:
-
-```
-Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
-```
-
-Spend 1-2 minutes investigating the relevant code:
-- Read the file(s) involved
-- Draw a quick ASCII diagram if it helps
-- Note any considerations
-
-```
-## Quick Exploration
-
-[Your brief analysis—what you found, any considerations]
-
-┌─────────────────────────────────────────┐
-│   [Optional: ASCII diagram if helpful]  │
-└─────────────────────────────────────────┘
-
-Explore mode (`/opsx:explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
-
-Now let's create a change to hold our work.
-```
-
-**PAUSE** - Wait for user acknowledgment before proceeding.
-
----
-
-## Phase 4: Create the Change
-
-**EXPLAIN:**
-```
-## Creating a Change
-
-A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
-
-Let me create one for our task.
-```
-
-**DO:** Create the change with a derived kebab-case name:
-```bash
-openspec new change "<derived-name>"
-```
-
-**SHOW:**
-```
-Created: <changeRoot from status JSON>
-
-The folder structure:
-```
-<changeRoot>/
-├── proposal.md    ← Why we're doing this (empty, we'll fill it)
-├── design.md      ← How we'll build it (empty)
-├── specs/         ← Detailed requirements (empty)
-└── tasks.md       ← Implementation checklist (empty)
-```
-
-Now let's fill in the first artifact—the proposal.
-```
-
----
-
-## Phase 5: Proposal
-
-**EXPLAIN:**
-```
-## The Proposal
-
-The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
-
-I'll draft one based on our task.
-```
-
-**DO:** Draft the proposal content (don't save yet):
-
-```
-Here's a draft proposal:
-
----
-
-## Why
-
-[1-2 sentences explaining the problem/opportunity]
-
-## What Changes
-
-[Bullet points of what will be different]
-
-## Capabilities
-
-### New Capabilities
-- `<capability-name>`: [brief description]
-
-### Modified Capabilities
-<!-- If modifying existing behavior -->
-
-## Impact
-
-- `src/path/to/file.ts`: [what changes]
-- [other files if applicable]
-
----
-
-Does this capture the intent? I can adjust before we save it.
-```
-
-**PAUSE** - Wait for user approval/feedback.
-
-After approval, save the proposal:
-```bash
-openspec instructions proposal --change "<name>" --json
-```
-Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
-
-```
-Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
-
-Next up: specs.
-```
-
----
-
-## Phase 6: Specs
-
-**EXPLAIN:**
-```
-## Specs
-
-Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
-
-For a small task like this, we might only need one spec file.
-```
-
-**DO:** Resolve where the spec file should be created:
-```bash
-openspec instructions specs --change "<name>" --json
-# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
-```
-
-Draft the spec content:
-
-```
-Here's the spec:
-
----
-
-## ADDED Requirements
-
-### Requirement: <Name>
-
-<Description of what the system should do>
-
-#### Scenario: <Scenario name>
-
-- **WHEN** <trigger condition>
-- **THEN** <expected outcome>
-- **AND** <additional outcome if needed>
-
----
-
-This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
-```
-
-Save to the concrete file path chosen from `resolvedOutputPath`.
-
----
-
-## Phase 7: Design
-
-**EXPLAIN:**
-```
-## Design
-
-The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
-
-For small changes, this might be brief. That's fine—not every change needs deep design discussion.
-```
-
-**DO:** Draft design.md:
-
-```
-Here's the design:
-
----
-
-## Context
-
-[Brief context about the current state]
-
-## Goals / Non-Goals
-
-**Goals:**
-- [What we're trying to achieve]
-
-**Non-Goals:**
-- [What's explicitly out of scope]
-
-## Decisions
-
-### Decision 1: [Key decision]
-
-[Explanation of approach and rationale]
-
----
-
-For a small task, this captures the key decisions without over-engineering.
-```
-
-Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
-
----
-
-## Phase 8: Tasks
-
-**EXPLAIN:**
-```
-## Tasks
-
-Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
-
-These should be small, clear, and in logical order.
-```
-
-**DO:** Generate tasks based on specs and design:
-
-```
-Here are the implementation tasks:
-
----
-
-## 1. [Category or file]
-
-- [ ] 1.1 [Specific task]
-- [ ] 1.2 [Specific task]
-
-## 2. Verify
-
-- [ ] 2.1 [Verification step]
-
----
-
-Each checkbox becomes a unit of work in the apply phase. Ready to implement?
-```
-
-**PAUSE** - Wait for user to confirm they're ready to implement.
-
-Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
-
----
-
-## Phase 9: Apply (Implementation)
-
-**EXPLAIN:**
-```
-## Implementation
-
-Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
-```
-
-**DO:** For each task:
-
-1. Announce: "Working on task N: [description]"
-2. Implement the change in the codebase
-3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
-4. Mark complete in tasks.md: `- [ ]` → `- [x]`
-5. Brief status: "✓ Task N complete"
-
-Keep narration light—don't over-explain every line of code.
-
-After all tasks:
-
-```
-## Implementation Complete
-
-All tasks done:
-- [x] Task 1
-- [x] Task 2
-- [x] ...
-
-The change is implemented! One more step—let's archive it.
-```
-
----
-
-## Phase 10: Archive
-
-**EXPLAIN:**
-```
-## Archiving
-
-When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
-
-Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
-```
-
-**DO:**
-```bash
-openspec archive "<name>"
-```
-
-**SHOW:**
-```
-Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
-
-The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
-```
-
----
-
-## Phase 11: Recap & Next Steps
-
-```
-## Congratulations!
-
-You just completed a full OpenSpec cycle:
-
-1. **Explore** - Thought through the problem
-2. **New** - Created a change container
-3. **Proposal** - Captured WHY
-4. **Specs** - Defined WHAT in detail
-5. **Design** - Decided HOW
-6. **Tasks** - Broke it into steps
-7. **Apply** - Implemented the work
-8. **Archive** - Preserved the record
-
-This same rhythm works for any size change—a small fix or a major feature.
-
----
-
-## Command Reference
-
-**Core workflow:**
-
- | Command           | What it does                               |
- |-------------------|--------------------------------------------|
- | `/opsx:propose` | Create a change and generate all artifacts |
- | `/opsx:explore` | Think through problems before/during work  |
- | `/opsx:apply`   | Implement tasks from a change              |
- | `/opsx:archive` | Archive a completed change                 |
-
-**Additional commands:**
-
- | Command            | What it does                                             |
- |--------------------|----------------------------------------------------------|
- | `/opsx:new`      | Start a new change, step through artifacts one at a time |
- | `/opsx:continue` | Continue working on an existing change                   |
- | `/opsx:ff`       | Fast-forward: create all artifacts at once               |
- | `/opsx:verify`   | Verify implementation matches artifacts                  |
-
----
-
-## What's Next?
-
-Try `/opsx:propose` on something you actually want to build. You've got the rhythm now!
-```
-
----
-
-## Graceful Exit Handling
-
-### User wants to stop mid-way
-
-If the user says they need to stop, want to pause, or seem disengaged:
-
-```
-No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
-
-To pick up where we left off later:
-- `/opsx:continue <name>` - Resume artifact creation
-- `/opsx:apply <name>` - Jump to implementation (if tasks exist)
-
-The work won't be lost. Come back whenever you're ready.
-```
-
-Exit gracefully without pressure.
-
-### User just wants command reference
-
-If the user says they just want to see the commands or skip the tutorial:
-
-```
-## OpenSpec Quick Reference
-
-**Core workflow:**
-
- | Command                  | What it does                               |
- |--------------------------|--------------------------------------------|
- | `/opsx:propose <name>` | Create a change and generate all artifacts |
- | `/opsx:explore`        | Think through problems (no code changes)   |
- | `/opsx:apply <name>`   | Implement tasks                            |
- | `/opsx:archive <name>` | Archive when done                          |
-
-**Additional commands:**
-
- | Command                   | What it does                        |
- |---------------------------|-------------------------------------|
- | `/opsx:new <name>`      | Start a new change, step by step    |
- | `/opsx:continue <name>` | Continue an existing change         |
- | `/opsx:ff <name>`       | Fast-forward: all artifacts at once |
- | `/opsx:verify <name>`   | Verify implementation               |
-
-Try `/opsx:propose` to start your first change.
-```
-
-Exit gracefully.
-
----
-
-## Guardrails
-
-- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
-- **Keep narration light** during implementation—teach without lecturing
-- **Don't skip phases** even if the change is small—the goal is teaching the workflow
-- **Pause for acknowledgment** at marked points, but don't over-pause
-- **Handle exits gracefully**—never pressure the user to continue
-- **Use real codebase tasks**—don't simulate or use fake examples
-- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.github/skills/openspec-propose/SKILL.md b/.github/skills/openspec-propose/SKILL.md
deleted file mode 100644
index 716d2d3d..00000000
--- a/.github/skills/openspec-propose/SKILL.md
+++ /dev/null
@@ -1,114 +0,0 @@
----
-name: openspec-propose
-description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
-
-**Steps**
-
-1. **If no clear input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/.github/skills/openspec-sync-specs/SKILL.md b/.github/skills/openspec-sync-specs/SKILL.md
deleted file mode 100644
index b6aad67a..00000000
--- a/.github/skills/openspec-sync-specs/SKILL.md
+++ /dev/null
@@ -1,148 +0,0 @@
----
-name: openspec-sync-specs
-description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/.github/skills/openspec-update-change/SKILL.md b/.github/skills/openspec-update-change/SKILL.md
deleted file mode 100644
index 3e1c931d..00000000
--- a/.github/skills/openspec-update-change/SKILL.md
+++ /dev/null
@@ -1,86 +0,0 @@
----
-name: openspec-update-change
-description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.github/skills/openspec-verify-change/SKILL.md b/.github/skills/openspec-verify-change/SKILL.md
deleted file mode 100644
index b6557c6f..00000000
--- a/.github/skills/openspec-verify-change/SKILL.md
+++ /dev/null
@@ -1,172 +0,0 @@
----
-name: openspec-verify-change
-description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/.opencode/commands/opsx-apply.md b/.opencode/commands/opsx-apply.md
deleted file mode 100644
index f5c574ff..00000000
--- a/.opencode/commands/opsx-apply.md
+++ /dev/null
@@ -1,152 +0,0 @@
----
-description: Implement tasks from an OpenSpec change (Experimental)
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name (e.g., `/opsx-apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx-apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx-continue`
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! You can archive this change with `/opsx-archive`.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.opencode/commands/opsx-archive.md b/.opencode/commands/opsx-archive.md
deleted file mode 100644
index 7f3a696f..00000000
--- a/.opencode/commands/opsx-archive.md
+++ /dev/null
@@ -1,157 +0,0 @@
----
-description: Archive a completed change in the experimental workflow
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx-archive` (e.g., `/opsx-archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Spec sync status (synced / sync skipped / no delta specs)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success (No Delta Specs)**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** No delta specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success With Warnings**
-
-```
-## Archive Complete (with warnings)
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** Sync skipped (user chose to skip)
-
-**Warnings:**
-- Archived with 2 incomplete artifacts
-- Archived with 3 incomplete tasks
-- Delta spec sync was skipped (user chose to skip)
-
-Review the archive if this was not intentional.
-```
-
-**Output On Error (Archive Exists)**
-
-```
-## Archive Failed
-
-**Change:** <change-name>
-**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-
-Target archive directory already exists.
-
-**Options:**
-1. Rename the existing archive
-2. Delete the existing archive if it's a duplicate
-3. Wait until a different date to archive
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.opencode/commands/opsx-explore.md b/.opencode/commands/opsx-explore.md
deleted file mode 100644
index 928233b4..00000000
--- a/.opencode/commands/opsx-explore.md
+++ /dev/null
@@ -1,171 +0,0 @@
----
-description: Enter explore mode - think through ideas, investigate problems, clarify requirements
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx-explore` is whatever the user wants to think about. Could be:
-- A vague idea: "real-time collaboration"
-- A specific problem: "the auth system is getting unwieldy"
-- A change name: "add-dark-mode" (to explore in context of that change)
-- A comparison: "postgres vs sqlite for this"
-- Nothing (just enter explore mode)
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-If the user mentioned a specific change name, read its artifacts for context.
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/.opencode/commands/opsx-onboard.md b/.opencode/commands/opsx-onboard.md
deleted file mode 100644
index 3ba27d12..00000000
--- a/.opencode/commands/opsx-onboard.md
+++ /dev/null
@@ -1,547 +0,0 @@
----
-description: Guided onboarding - walk through a complete OpenSpec workflow cycle with narration
----
-
-Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## Preflight
-
-Before starting, check if the OpenSpec CLI is installed:
-
-```bash
-# Unix/macOS
-openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
-# Windows (PowerShell)
-# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
-```
-
-**If CLI not installed:**
-> OpenSpec CLI is not installed. Install it first, then come back to `/opsx-onboard`.
-
-Stop here if not installed.
-
----
-
-## Phase 1: Welcome
-
-Display:
-
-```
-## Welcome to OpenSpec!
-
-I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
-
-**What we'll do:**
-1. Pick a small, real task in your codebase
-2. Explore the problem briefly
-3. Create a change (the container for our work)
-4. Build the artifacts: proposal → specs → design → tasks
-5. Implement the tasks
-6. Archive the completed change
-
-**Time:** ~15-20 minutes
-
-Let's start by finding something to work on.
-```
-
----
-
-## Phase 2: Task Selection
-
-### Codebase Analysis
-
-Scan the codebase for small improvement opportunities. Look for:
-
-1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
-2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
-3. **Functions without tests** - Cross-reference `src/` with test directories
-4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
-5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
-6. **Missing validation** - User input handlers without validation
-
-Also check recent git activity:
-```bash
-# Unix/macOS
-git log --oneline -10 2>/dev/null || echo "No git history"
-# Windows (PowerShell)
-# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
-```
-
-### Present Suggestions
-
-From your analysis, present 3-4 specific suggestions:
-
-```
-## Task Suggestions
-
-Based on scanning your codebase, here are some good starter tasks:
-
-**1. [Most promising task]**
-   Location: `src/path/to/file.ts:42`
-   Scope: ~1-2 files, ~20-30 lines
-   Why it's good: [brief reason]
-
-**2. [Second task]**
-   Location: `src/another/file.ts`
-   Scope: ~1 file, ~15 lines
-   Why it's good: [brief reason]
-
-**3. [Third task]**
-   Location: [location]
-   Scope: [estimate]
-   Why it's good: [brief reason]
-
-**4. Something else?**
-   Tell me what you'd like to work on.
-
-Which task interests you? (Pick a number or describe your own)
-```
-
-**If nothing found:** Fall back to asking what the user wants to build:
-> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
-
-### Scope Guardrail
-
-If the user picks or describes something too large (major feature, multi-day work):
-
-```
-That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
-
-For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
-
-**Options:**
-1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
-2. **Pick something else** - One of the other suggestions, or a different small task?
-3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
-
-What would you prefer?
-```
-
-Let the user override if they insist—this is a soft guardrail.
-
----
-
-## Phase 3: Explore Demo
-
-Once a task is selected, briefly demonstrate explore mode:
-
-```
-Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
-```
-
-Spend 1-2 minutes investigating the relevant code:
-- Read the file(s) involved
-- Draw a quick ASCII diagram if it helps
-- Note any considerations
-
-```
-## Quick Exploration
-
-[Your brief analysis—what you found, any considerations]
-
-┌─────────────────────────────────────────┐
-│   [Optional: ASCII diagram if helpful]  │
-└─────────────────────────────────────────┘
-
-Explore mode (`/opsx-explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
-
-Now let's create a change to hold our work.
-```
-
-**PAUSE** - Wait for user acknowledgment before proceeding.
-
----
-
-## Phase 4: Create the Change
-
-**EXPLAIN:**
-```
-## Creating a Change
-
-A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
-
-Let me create one for our task.
-```
-
-**DO:** Create the change with a derived kebab-case name:
-```bash
-openspec new change "<derived-name>"
-```
-
-**SHOW:**
-```
-Created: <changeRoot from status JSON>
-
-The folder structure:
-```
-<changeRoot>/
-├── proposal.md    ← Why we're doing this (empty, we'll fill it)
-├── design.md      ← How we'll build it (empty)
-├── specs/         ← Detailed requirements (empty)
-└── tasks.md       ← Implementation checklist (empty)
-```
-
-Now let's fill in the first artifact—the proposal.
-```
-
----
-
-## Phase 5: Proposal
-
-**EXPLAIN:**
-```
-## The Proposal
-
-The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
-
-I'll draft one based on our task.
-```
-
-**DO:** Draft the proposal content (don't save yet):
-
-```
-Here's a draft proposal:
-
----
-
-## Why
-
-[1-2 sentences explaining the problem/opportunity]
-
-## What Changes
-
-[Bullet points of what will be different]
-
-## Capabilities
-
-### New Capabilities
-- `<capability-name>`: [brief description]
-
-### Modified Capabilities
-<!-- If modifying existing behavior -->
-
-## Impact
-
-- `src/path/to/file.ts`: [what changes]
-- [other files if applicable]
-
----
-
-Does this capture the intent? I can adjust before we save it.
-```
-
-**PAUSE** - Wait for user approval/feedback.
-
-After approval, save the proposal:
-```bash
-openspec instructions proposal --change "<name>" --json
-```
-Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
-
-```
-Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
-
-Next up: specs.
-```
-
----
-
-## Phase 6: Specs
-
-**EXPLAIN:**
-```
-## Specs
-
-Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
-
-For a small task like this, we might only need one spec file.
-```
-
-**DO:** Resolve where the spec file should be created:
-```bash
-openspec instructions specs --change "<name>" --json
-# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
-```
-
-Draft the spec content:
-
-```
-Here's the spec:
-
----
-
-## ADDED Requirements
-
-### Requirement: <Name>
-
-<Description of what the system should do>
-
-#### Scenario: <Scenario name>
-
-- **WHEN** <trigger condition>
-- **THEN** <expected outcome>
-- **AND** <additional outcome if needed>
-
----
-
-This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
-```
-
-Save to the concrete file path chosen from `resolvedOutputPath`.
-
----
-
-## Phase 7: Design
-
-**EXPLAIN:**
-```
-## Design
-
-The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
-
-For small changes, this might be brief. That's fine—not every change needs deep design discussion.
-```
-
-**DO:** Draft design.md:
-
-```
-Here's the design:
-
----
-
-## Context
-
-[Brief context about the current state]
-
-## Goals / Non-Goals
-
-**Goals:**
-- [What we're trying to achieve]
-
-**Non-Goals:**
-- [What's explicitly out of scope]
-
-## Decisions
-
-### Decision 1: [Key decision]
-
-[Explanation of approach and rationale]
-
----
-
-For a small task, this captures the key decisions without over-engineering.
-```
-
-Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
-
----
-
-## Phase 8: Tasks
-
-**EXPLAIN:**
-```
-## Tasks
-
-Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
-
-These should be small, clear, and in logical order.
-```
-
-**DO:** Generate tasks based on specs and design:
-
-```
-Here are the implementation tasks:
-
----
-
-## 1. [Category or file]
-
-- [ ] 1.1 [Specific task]
-- [ ] 1.2 [Specific task]
-
-## 2. Verify
-
-- [ ] 2.1 [Verification step]
-
----
-
-Each checkbox becomes a unit of work in the apply phase. Ready to implement?
-```
-
-**PAUSE** - Wait for user to confirm they're ready to implement.
-
-Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
-
----
-
-## Phase 9: Apply (Implementation)
-
-**EXPLAIN:**
-```
-## Implementation
-
-Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
-```
-
-**DO:** For each task:
-
-1. Announce: "Working on task N: [description]"
-2. Implement the change in the codebase
-3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
-4. Mark complete in tasks.md: `- [ ]` → `- [x]`
-5. Brief status: "✓ Task N complete"
-
-Keep narration light—don't over-explain every line of code.
-
-After all tasks:
-
-```
-## Implementation Complete
-
-All tasks done:
-- [x] Task 1
-- [x] Task 2
-- [x] ...
-
-The change is implemented! One more step—let's archive it.
-```
-
----
-
-## Phase 10: Archive
-
-**EXPLAIN:**
-```
-## Archiving
-
-When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
-
-Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
-```
-
-**DO:**
-```bash
-openspec archive "<name>"
-```
-
-**SHOW:**
-```
-Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
-
-The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
-```
-
----
-
-## Phase 11: Recap & Next Steps
-
-```
-## Congratulations!
-
-You just completed a full OpenSpec cycle:
-
-1. **Explore** - Thought through the problem
-2. **New** - Created a change container
-3. **Proposal** - Captured WHY
-4. **Specs** - Defined WHAT in detail
-5. **Design** - Decided HOW
-6. **Tasks** - Broke it into steps
-7. **Apply** - Implemented the work
-8. **Archive** - Preserved the record
-
-This same rhythm works for any size change—a small fix or a major feature.
-
----
-
-## Command Reference
-
-**Core workflow:**
-
- | Command           | What it does                               |
- |-------------------|--------------------------------------------|
- | `/opsx-propose` | Create a change and generate all artifacts |
- | `/opsx-explore` | Think through problems before/during work  |
- | `/opsx-apply`   | Implement tasks from a change              |
- | `/opsx-archive` | Archive a completed change                 |
-
-**Additional commands:**
-
- | Command            | What it does                                             |
- |--------------------|----------------------------------------------------------|
- | `/opsx-new`      | Start a new change, step through artifacts one at a time |
- | `/opsx-continue` | Continue working on an existing change                   |
- | `/opsx-ff`       | Fast-forward: create all artifacts at once               |
- | `/opsx-verify`   | Verify implementation matches artifacts                  |
-
----
-
-## What's Next?
-
-Try `/opsx-propose` on something you actually want to build. You've got the rhythm now!
-```
-
----
-
-## Graceful Exit Handling
-
-### User wants to stop mid-way
-
-If the user says they need to stop, want to pause, or seem disengaged:
-
-```
-No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
-
-To pick up where we left off later:
-- `/opsx-continue <name>` - Resume artifact creation
-- `/opsx-apply <name>` - Jump to implementation (if tasks exist)
-
-The work won't be lost. Come back whenever you're ready.
-```
-
-Exit gracefully without pressure.
-
-### User just wants command reference
-
-If the user says they just want to see the commands or skip the tutorial:
-
-```
-## OpenSpec Quick Reference
-
-**Core workflow:**
-
- | Command                  | What it does                               |
- |--------------------------|--------------------------------------------|
- | `/opsx-propose <name>` | Create a change and generate all artifacts |
- | `/opsx-explore`        | Think through problems (no code changes)   |
- | `/opsx-apply <name>`   | Implement tasks                            |
- | `/opsx-archive <name>` | Archive when done                          |
-
-**Additional commands:**
-
- | Command                   | What it does                        |
- |---------------------------|-------------------------------------|
- | `/opsx-new <name>`      | Start a new change, step by step    |
- | `/opsx-continue <name>` | Continue an existing change         |
- | `/opsx-ff <name>`       | Fast-forward: all artifacts at once |
- | `/opsx-verify <name>`   | Verify implementation               |
-
-Try `/opsx-propose` to start your first change.
-```
-
-Exit gracefully.
-
----
-
-## Guardrails
-
-- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
-- **Keep narration light** during implementation—teach without lecturing
-- **Don't skip phases** even if the change is small—the goal is teaching the workflow
-- **Pause for acknowledgment** at marked points, but don't over-pause
-- **Handle exits gracefully**—never pressure the user to continue
-- **Use real codebase tasks**—don't simulate or use fake examples
-- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.opencode/commands/opsx-propose.md b/.opencode/commands/opsx-propose.md
deleted file mode 100644
index 6b498c26..00000000
--- a/.opencode/commands/opsx-propose.md
+++ /dev/null
@@ -1,106 +0,0 @@
----
-description: Propose a new change - create it and generate all artifacts in one step
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx-apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx-propose` is the change name (kebab-case), OR a description of what the user wants to build.
-
-**Steps**
-
-1. **If no input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx-apply` to start implementing."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/.opencode/commands/opsx-sync.md b/.opencode/commands/opsx-sync.md
deleted file mode 100644
index 5f1b7bb4..00000000
--- a/.opencode/commands/opsx-sync.md
+++ /dev/null
@@ -1,140 +0,0 @@
----
-description: Sync delta specs from a change to main specs
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx-sync` (e.g., `/opsx-sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/.opencode/commands/opsx-update.md b/.opencode/commands/opsx-update.md
deleted file mode 100644
index 7327a92f..00000000
--- a/.opencode/commands/opsx-update.md
+++ /dev/null
@@ -1,78 +0,0 @@
----
-description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx-update` (e.g., `/opsx-update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx-continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx-continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx-apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx-archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx-continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx-apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx-continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx-new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.opencode/commands/opsx-verify.md b/.opencode/commands/opsx-verify.md
deleted file mode 100644
index de6d1450..00000000
--- a/.opencode/commands/opsx-verify.md
+++ /dev/null
@@ -1,164 +0,0 @@
----
-description: Verify implementation matches change artifacts before archiving
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx-verify` (e.g., `/opsx-verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/.opencode/skills/openspec-apply-change/SKILL.md b/.opencode/skills/openspec-apply-change/SKILL.md
deleted file mode 100644
index e44e01f9..00000000
--- a/.opencode/skills/openspec-apply-change/SKILL.md
+++ /dev/null
@@ -1,160 +0,0 @@
----
-name: openspec-apply-change
-description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx-apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! Ready to archive this change.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/.opencode/skills/openspec-archive-change/SKILL.md b/.opencode/skills/openspec-archive-change/SKILL.md
deleted file mode 100644
index c0c169d1..00000000
--- a/.opencode/skills/openspec-archive-change/SKILL.md
+++ /dev/null
@@ -1,118 +0,0 @@
----
-name: openspec-archive-change
-description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Whether specs were synced (if applicable)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
-
-All artifacts complete. All tasks complete.
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use openspec-sync-specs approach (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/.opencode/skills/openspec-explore/SKILL.md b/.opencode/skills/openspec-explore/SKILL.md
deleted file mode 100644
index 894ef32e..00000000
--- a/.opencode/skills/openspec-explore/SKILL.md
+++ /dev/null
@@ -1,290 +0,0 @@
----
-name: openspec-explore
-description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Handling Different Entry Points
-
-**User brings a vague idea:**
-```
-User: I'm thinking about adding real-time collaboration
-
-You: Real-time collab is a big space. Let me think about this...
-
-      COLLABORATION SPECTRUM
-      ════════════════════════════════════════════
-
-      Awareness          Coordination         Sync
-          │                   │                 │
-          ▼                   ▼                 ▼
-      ┌────────┐         ┌────────┐        ┌────────┐
-      │Presence│         │Cursors │        │  CRDT  │
-      │  "3    │         │ Multi  │        │Conflict│
-      │online" │         │ select │        │  free  │
-      └────────┘         └────────┘        └────────┘
-          │                   │                 │
-       trivial            moderate           complex
-
-      Where's your head at?
-```
-
-**User brings a specific problem:**
-```
-User: The auth system is a mess
-
-You: [reads codebase]
-
-     ┌─────────────────────────────────────────────┐
-     │              CURRENT AUTH FLOW              │
-     └─────────────────────────────────────────────┘
-                        │
-          ┌─────────────┼─────────────┐
-          ▼             ▼             ▼
-     ┌─────────┐  ┌─────────┐  ┌─────────┐
-     │ Google  │  │  GitHub │  │  Email  │
-     │  OAuth  │  │  OAuth  │  │  Magic  │
-     └────┬────┘  └────┬────┘  └────┬────┘
-          │             │             │
-          └─────────────┼─────────────┘
-                        ▼
-                  ┌───────────┐
-                  │  Session  │
-                  └─────┬─────┘
-                        │
-                        ▼
-                  ┌───────────┐
-                  │   Perms   │
-                  └───────────┘
-
-     I see three tangles. Which one's burning?
-```
-
-**User is stuck mid-implementation:**
-```
-User: /opsx-explore add-auth-system
-      The OAuth integration is more complex than expected
-
-You: [reads change artifacts]
-
-     You're on task 4: "Implement OAuth flow"
-
-     Let me trace what's involved...
-
-     [draws diagram, explores options, suggests paths]
-
-     Want to update the design to reflect this?
-     Or add a spike task to investigate?
-```
-
-**User wants to compare options:**
-```
-User: Should we use Postgres or SQLite?
-
-You: Generic answer is boring. What's the context?
-
-User: A CLI tool that tracks local dev environments
-
-You: That changes everything.
-
-     ┌─────────────────────────────────────────────────┐
-     │          CLI TOOL DATA STORAGE                  │
-     └─────────────────────────────────────────────────┘
-
-     Key constraints:
-     • No daemon running
-     • Must work offline
-     • Single user
-
-                  SQLite          Postgres
-     Deployment   embedded ✓      needs server ✗
-     Offline      yes ✓           no ✗
-     Single file  yes ✓           no ✗
-
-     SQLite. Not even close.
-
-     Unless... is there a sync component?
-```
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When it feels like things are crystallizing, you might summarize:
-
-```
-## What We Figured Out
-
-**The problem**: [crystallized understanding]
-
-**The approach**: [if one emerged]
-
-**Open questions**: [if any remain]
-
-**Next steps** (if ready):
-- Create a change proposal
-- Keep exploring: just keep talking
-```
-
-But this summary is optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/.opencode/skills/openspec-onboard/SKILL.md b/.opencode/skills/openspec-onboard/SKILL.md
deleted file mode 100644
index 52c4228f..00000000
--- a/.opencode/skills/openspec-onboard/SKILL.md
+++ /dev/null
@@ -1,555 +0,0 @@
----
-name: openspec-onboard
-description: Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Guide the user through their first complete OpenSpec workflow cycle. This is a teaching experience—you'll do real work in their codebase while explaining each step.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## Preflight
-
-Before starting, check if the OpenSpec CLI is installed:
-
-```bash
-# Unix/macOS
-openspec --version 2>&1 || echo "CLI_NOT_INSTALLED"
-# Windows (PowerShell)
-# if (Get-Command openspec -ErrorAction SilentlyContinue) { openspec --version } else { echo "CLI_NOT_INSTALLED" }
-```
-
-**If CLI not installed:**
-> OpenSpec CLI is not installed. Install it first, then come back to `/opsx-onboard`.
-
-Stop here if not installed.
-
----
-
-## Phase 1: Welcome
-
-Display:
-
-```
-## Welcome to OpenSpec!
-
-I'll walk you through a complete change cycle—from idea to implementation—using a real task in your codebase. Along the way, you'll learn the workflow by doing it.
-
-**What we'll do:**
-1. Pick a small, real task in your codebase
-2. Explore the problem briefly
-3. Create a change (the container for our work)
-4. Build the artifacts: proposal → specs → design → tasks
-5. Implement the tasks
-6. Archive the completed change
-
-**Time:** ~15-20 minutes
-
-Let's start by finding something to work on.
-```
-
----
-
-## Phase 2: Task Selection
-
-### Codebase Analysis
-
-Scan the codebase for small improvement opportunities. Look for:
-
-1. **TODO/FIXME comments** - Search for `TODO`, `FIXME`, `HACK`, `XXX` in code files
-2. **Missing error handling** - `catch` blocks that swallow errors, risky operations without try-catch
-3. **Functions without tests** - Cross-reference `src/` with test directories
-4. **Type issues** - `any` types in TypeScript files (`: any`, `as any`)
-5. **Debug artifacts** - `console.log`, `console.debug`, `debugger` statements in non-debug code
-6. **Missing validation** - User input handlers without validation
-
-Also check recent git activity:
-```bash
-# Unix/macOS
-git log --oneline -10 2>/dev/null || echo "No git history"
-# Windows (PowerShell)
-# git log --oneline -10 2>$null; if ($LASTEXITCODE -ne 0) { echo "No git history" }
-```
-
-### Present Suggestions
-
-From your analysis, present 3-4 specific suggestions:
-
-```
-## Task Suggestions
-
-Based on scanning your codebase, here are some good starter tasks:
-
-**1. [Most promising task]**
-   Location: `src/path/to/file.ts:42`
-   Scope: ~1-2 files, ~20-30 lines
-   Why it's good: [brief reason]
-
-**2. [Second task]**
-   Location: `src/another/file.ts`
-   Scope: ~1 file, ~15 lines
-   Why it's good: [brief reason]
-
-**3. [Third task]**
-   Location: [location]
-   Scope: [estimate]
-   Why it's good: [brief reason]
-
-**4. Something else?**
-   Tell me what you'd like to work on.
-
-Which task interests you? (Pick a number or describe your own)
-```
-
-**If nothing found:** Fall back to asking what the user wants to build:
-> I didn't find obvious quick wins in your codebase. What's something small you've been meaning to add or fix?
-
-### Scope Guardrail
-
-If the user picks or describes something too large (major feature, multi-day work):
-
-```
-That's a valuable task, but it's probably larger than ideal for your first OpenSpec run-through.
-
-For learning the workflow, smaller is better—it lets you see the full cycle without getting stuck in implementation details.
-
-**Options:**
-1. **Slice it smaller** - What's the smallest useful piece of [their task]? Maybe just [specific slice]?
-2. **Pick something else** - One of the other suggestions, or a different small task?
-3. **Do it anyway** - If you really want to tackle this, we can. Just know it'll take longer.
-
-What would you prefer?
-```
-
-Let the user override if they insist—this is a soft guardrail.
-
----
-
-## Phase 3: Explore Demo
-
-Once a task is selected, briefly demonstrate explore mode:
-
-```
-Before we create a change, let me quickly show you **explore mode**—it's how you think through problems before committing to a direction.
-```
-
-Spend 1-2 minutes investigating the relevant code:
-- Read the file(s) involved
-- Draw a quick ASCII diagram if it helps
-- Note any considerations
-
-```
-## Quick Exploration
-
-[Your brief analysis—what you found, any considerations]
-
-┌─────────────────────────────────────────┐
-│   [Optional: ASCII diagram if helpful]  │
-└─────────────────────────────────────────┘
-
-Explore mode (`/opsx-explore`) is for this kind of thinking—investigating before implementing. You can use it anytime you need to think through a problem.
-
-Now let's create a change to hold our work.
-```
-
-**PAUSE** - Wait for user acknowledgment before proceeding.
-
----
-
-## Phase 4: Create the Change
-
-**EXPLAIN:**
-```
-## Creating a Change
-
-A "change" in OpenSpec is a container for all the thinking and planning around a piece of work. It lives at the `changeRoot` reported by `openspec status --change "<name>" --json` and holds your artifacts—proposal, specs, design, tasks.
-
-Let me create one for our task.
-```
-
-**DO:** Create the change with a derived kebab-case name:
-```bash
-openspec new change "<derived-name>"
-```
-
-**SHOW:**
-```
-Created: <changeRoot from status JSON>
-
-The folder structure:
-```
-<changeRoot>/
-├── proposal.md    ← Why we're doing this (empty, we'll fill it)
-├── design.md      ← How we'll build it (empty)
-├── specs/         ← Detailed requirements (empty)
-└── tasks.md       ← Implementation checklist (empty)
-```
-
-Now let's fill in the first artifact—the proposal.
-```
-
----
-
-## Phase 5: Proposal
-
-**EXPLAIN:**
-```
-## The Proposal
-
-The proposal captures **why** we're making this change and **what** it involves at a high level. It's the "elevator pitch" for the work.
-
-I'll draft one based on our task.
-```
-
-**DO:** Draft the proposal content (don't save yet):
-
-```
-Here's a draft proposal:
-
----
-
-## Why
-
-[1-2 sentences explaining the problem/opportunity]
-
-## What Changes
-
-[Bullet points of what will be different]
-
-## Capabilities
-
-### New Capabilities
-- `<capability-name>`: [brief description]
-
-### Modified Capabilities
-<!-- If modifying existing behavior -->
-
-## Impact
-
-- `src/path/to/file.ts`: [what changes]
-- [other files if applicable]
-
----
-
-Does this capture the intent? I can adjust before we save it.
-```
-
-**PAUSE** - Wait for user approval/feedback.
-
-After approval, save the proposal:
-```bash
-openspec instructions proposal --change "<name>" --json
-```
-Then write the content to the `resolvedOutputPath` from `openspec instructions proposal --change "<name>" --json`.
-
-```
-Proposal saved. This is your "why" document—you can always come back and refine it as understanding evolves.
-
-Next up: specs.
-```
-
----
-
-## Phase 6: Specs
-
-**EXPLAIN:**
-```
-## Specs
-
-Specs define **what** we're building in precise, testable terms. They use a requirement/scenario format that makes expected behavior crystal clear.
-
-For a small task like this, we might only need one spec file.
-```
-
-**DO:** Resolve where the spec file should be created:
-```bash
-openspec instructions specs --change "<name>" --json
-# Use resolvedOutputPath from the JSON. If it is a glob, choose the concrete file path using the schema instruction and the change's context.
-```
-
-Draft the spec content:
-
-```
-Here's the spec:
-
----
-
-## ADDED Requirements
-
-### Requirement: <Name>
-
-<Description of what the system should do>
-
-#### Scenario: <Scenario name>
-
-- **WHEN** <trigger condition>
-- **THEN** <expected outcome>
-- **AND** <additional outcome if needed>
-
----
-
-This format—WHEN/THEN/AND—makes requirements testable. You can literally read them as test cases.
-```
-
-Save to the concrete file path chosen from `resolvedOutputPath`.
-
----
-
-## Phase 7: Design
-
-**EXPLAIN:**
-```
-## Design
-
-The design captures **how** we'll build it—technical decisions, tradeoffs, approach.
-
-For small changes, this might be brief. That's fine—not every change needs deep design discussion.
-```
-
-**DO:** Draft design.md:
-
-```
-Here's the design:
-
----
-
-## Context
-
-[Brief context about the current state]
-
-## Goals / Non-Goals
-
-**Goals:**
-- [What we're trying to achieve]
-
-**Non-Goals:**
-- [What's explicitly out of scope]
-
-## Decisions
-
-### Decision 1: [Key decision]
-
-[Explanation of approach and rationale]
-
----
-
-For a small task, this captures the key decisions without over-engineering.
-```
-
-Save to the `resolvedOutputPath` from `openspec instructions design --change "<name>" --json`.
-
----
-
-## Phase 8: Tasks
-
-**EXPLAIN:**
-```
-## Tasks
-
-Finally, we break the work into implementation tasks—checkboxes that drive the apply phase.
-
-These should be small, clear, and in logical order.
-```
-
-**DO:** Generate tasks based on specs and design:
-
-```
-Here are the implementation tasks:
-
----
-
-## 1. [Category or file]
-
-- [ ] 1.1 [Specific task]
-- [ ] 1.2 [Specific task]
-
-## 2. Verify
-
-- [ ] 2.1 [Verification step]
-
----
-
-Each checkbox becomes a unit of work in the apply phase. Ready to implement?
-```
-
-**PAUSE** - Wait for user to confirm they're ready to implement.
-
-Save to the `resolvedOutputPath` from `openspec instructions tasks --change "<name>" --json`.
-
----
-
-## Phase 9: Apply (Implementation)
-
-**EXPLAIN:**
-```
-## Implementation
-
-Now we implement each task, checking them off as we go. I'll announce each one and occasionally note how the specs/design informed the approach.
-```
-
-**DO:** For each task:
-
-1. Announce: "Working on task N: [description]"
-2. Implement the change in the codebase
-3. Reference specs/design naturally: "The spec says X, so I'm doing Y"
-4. Mark complete in tasks.md: `- [ ]` → `- [x]`
-5. Brief status: "✓ Task N complete"
-
-Keep narration light—don't over-explain every line of code.
-
-After all tasks:
-
-```
-## Implementation Complete
-
-All tasks done:
-- [x] Task 1
-- [x] Task 2
-- [x] ...
-
-The change is implemented! One more step—let's archive it.
-```
-
----
-
-## Phase 10: Archive
-
-**EXPLAIN:**
-```
-## Archiving
-
-When a change is complete, we archive it. The archive path is derived from `planningHome.changesDir` and the date.
-
-Archived changes become your project's decision history—you can always find them later to understand why something was built a certain way.
-```
-
-**DO:**
-```bash
-openspec archive "<name>"
-```
-
-**SHOW:**
-```
-Archived to: `<planningHome.changesDir>/archive/YYYY-MM-DD-<name>/`
-
-The change is now part of your project's history. The code is in your codebase, the decision record is preserved.
-```
-
----
-
-## Phase 11: Recap & Next Steps
-
-```
-## Congratulations!
-
-You just completed a full OpenSpec cycle:
-
-1. **Explore** - Thought through the problem
-2. **New** - Created a change container
-3. **Proposal** - Captured WHY
-4. **Specs** - Defined WHAT in detail
-5. **Design** - Decided HOW
-6. **Tasks** - Broke it into steps
-7. **Apply** - Implemented the work
-8. **Archive** - Preserved the record
-
-This same rhythm works for any size change—a small fix or a major feature.
-
----
-
-## Command Reference
-
-**Core workflow:**
-
- | Command           | What it does                               |
- |-------------------|--------------------------------------------|
- | `/opsx-propose` | Create a change and generate all artifacts |
- | `/opsx-explore` | Think through problems before/during work  |
- | `/opsx-apply`   | Implement tasks from a change              |
- | `/opsx-archive` | Archive a completed change                 |
-
-**Additional commands:**
-
- | Command            | What it does                                             |
- |--------------------|----------------------------------------------------------|
- | `/opsx-new`      | Start a new change, step through artifacts one at a time |
- | `/opsx-continue` | Continue working on an existing change                   |
- | `/opsx-ff`       | Fast-forward: create all artifacts at once               |
- | `/opsx-verify`   | Verify implementation matches artifacts                  |
-
----
-
-## What's Next?
-
-Try `/opsx-propose` on something you actually want to build. You've got the rhythm now!
-```
-
----
-
-## Graceful Exit Handling
-
-### User wants to stop mid-way
-
-If the user says they need to stop, want to pause, or seem disengaged:
-
-```
-No problem! Your change is saved at the `changeRoot` reported by `openspec status --change "<name>" --json`.
-
-To pick up where we left off later:
-- `/opsx-continue <name>` - Resume artifact creation
-- `/opsx-apply <name>` - Jump to implementation (if tasks exist)
-
-The work won't be lost. Come back whenever you're ready.
-```
-
-Exit gracefully without pressure.
-
-### User just wants command reference
-
-If the user says they just want to see the commands or skip the tutorial:
-
-```
-## OpenSpec Quick Reference
-
-**Core workflow:**
-
- | Command                  | What it does                               |
- |--------------------------|--------------------------------------------|
- | `/opsx-propose <name>` | Create a change and generate all artifacts |
- | `/opsx-explore`        | Think through problems (no code changes)   |
- | `/opsx-apply <name>`   | Implement tasks                            |
- | `/opsx-archive <name>` | Archive when done                          |
-
-**Additional commands:**
-
- | Command                   | What it does                        |
- |---------------------------|-------------------------------------|
- | `/opsx-new <name>`      | Start a new change, step by step    |
- | `/opsx-continue <name>` | Continue an existing change         |
- | `/opsx-ff <name>`       | Fast-forward: all artifacts at once |
- | `/opsx-verify <name>`   | Verify implementation               |
-
-Try `/opsx-propose` to start your first change.
-```
-
-Exit gracefully.
-
----
-
-## Guardrails
-
-- **Follow the EXPLAIN → DO → SHOW → PAUSE pattern** at key transitions (after explore, after proposal draft, after tasks, after archive)
-- **Keep narration light** during implementation—teach without lecturing
-- **Don't skip phases** even if the change is small—the goal is teaching the workflow
-- **Pause for acknowledgment** at marked points, but don't over-pause
-- **Handle exits gracefully**—never pressure the user to continue
-- **Use real codebase tasks**—don't simulate or use fake examples
-- **Adjust scope gently**—guide toward smaller tasks but respect user choice
diff --git a/.opencode/skills/openspec-propose/SKILL.md b/.opencode/skills/openspec-propose/SKILL.md
deleted file mode 100644
index f0af3e15..00000000
--- a/.opencode/skills/openspec-propose/SKILL.md
+++ /dev/null
@@ -1,114 +0,0 @@
----
-name: openspec-propose
-description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx-apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
-
-**Steps**
-
-1. **If no clear input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx-apply` or ask me to implement to start working on the tasks."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/.opencode/skills/openspec-sync-specs/SKILL.md b/.opencode/skills/openspec-sync-specs/SKILL.md
deleted file mode 100644
index b6aad67a..00000000
--- a/.opencode/skills/openspec-sync-specs/SKILL.md
+++ /dev/null
@@ -1,148 +0,0 @@
----
-name: openspec-sync-specs
-description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/.opencode/skills/openspec-update-change/SKILL.md b/.opencode/skills/openspec-update-change/SKILL.md
deleted file mode 100644
index efa37f4c..00000000
--- a/.opencode/skills/openspec-update-change/SKILL.md
+++ /dev/null
@@ -1,86 +0,0 @@
----
-name: openspec-update-change
-description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx-continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx-continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx-apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx-archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx-continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx-apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx-continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx-new` (the "Update vs. Start Fresh" heuristic).
diff --git a/.opencode/skills/openspec-verify-change/SKILL.md b/.opencode/skills/openspec-verify-change/SKILL.md
deleted file mode 100644
index b6557c6f..00000000
--- a/.opencode/skills/openspec-verify-change/SKILL.md
+++ /dev/null
@@ -1,172 +0,0 @@
----
-name: openspec-verify-change
-description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"

From d3f311b523c773f7c30d172ae110bf8ed5a56cc4 Mon Sep 17 00:00:00 2001
From: gohabereg <gohabereg@gmail.com>
Date: Sat, 18 Jul 2026 17:06:00 +0100
Subject: [PATCH 4/5] Add CONTRIBUTING.md and remove stray packages/ openspec
 scaffolding

Documents setup, testing, and the OpenSpec-driven change workflow.
Also removes packages/.claude, packages/.github, and packages/openspec,
which were duplicate/misplaced files from running openspec tooling
with the wrong working directory.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
---
 CONTRIBUTING.md                               | 142 +++++++++
 packages/.claude/commands/opsx/apply.md       | 156 ----------
 packages/.claude/commands/opsx/archive.md     | 161 ----------
 packages/.claude/commands/opsx/explore.md     | 175 -----------
 packages/.claude/commands/opsx/propose.md     | 110 -------
 packages/.claude/commands/opsx/sync.md        | 144 ---------
 packages/.claude/commands/opsx/update.md      |  82 -----
 packages/.claude/commands/opsx/verify.md      | 168 ----------
 .../skills/openspec-apply-change/SKILL.md     | 160 ----------
 .../skills/openspec-archive-change/SKILL.md   | 118 -------
 .../.claude/skills/openspec-explore/SKILL.md  | 290 ------------------
 .../.claude/skills/openspec-propose/SKILL.md  | 114 -------
 .../skills/openspec-sync-specs/SKILL.md       | 148 ---------
 .../skills/openspec-update-change/SKILL.md    |  86 ------
 .../skills/openspec-verify-change/SKILL.md    | 172 -----------
 packages/.github/prompts/opsx-apply.prompt.md | 152 ---------
 .../.github/prompts/opsx-archive.prompt.md    | 157 ----------
 .../.github/prompts/opsx-explore.prompt.md    | 171 -----------
 .../.github/prompts/opsx-propose.prompt.md    | 106 -------
 packages/.github/prompts/opsx-sync.prompt.md  | 140 ---------
 .../.github/prompts/opsx-update.prompt.md     |  78 -----
 .../.github/prompts/opsx-verify.prompt.md     | 164 ----------
 .../skills/openspec-apply-change/SKILL.md     | 160 ----------
 .../skills/openspec-archive-change/SKILL.md   | 118 -------
 .../.github/skills/openspec-explore/SKILL.md  | 290 ------------------
 .../.github/skills/openspec-propose/SKILL.md  | 114 -------
 .../skills/openspec-sync-specs/SKILL.md       | 148 ---------
 .../skills/openspec-update-change/SKILL.md    |  86 ------
 .../skills/openspec-verify-change/SKILL.md    | 172 -----------
 packages/openspec/config.yaml                 |  20 --
 30 files changed, 142 insertions(+), 4160 deletions(-)
 create mode 100644 CONTRIBUTING.md
 delete mode 100644 packages/.claude/commands/opsx/apply.md
 delete mode 100644 packages/.claude/commands/opsx/archive.md
 delete mode 100644 packages/.claude/commands/opsx/explore.md
 delete mode 100644 packages/.claude/commands/opsx/propose.md
 delete mode 100644 packages/.claude/commands/opsx/sync.md
 delete mode 100644 packages/.claude/commands/opsx/update.md
 delete mode 100644 packages/.claude/commands/opsx/verify.md
 delete mode 100644 packages/.claude/skills/openspec-apply-change/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-archive-change/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-explore/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-propose/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-sync-specs/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-update-change/SKILL.md
 delete mode 100644 packages/.claude/skills/openspec-verify-change/SKILL.md
 delete mode 100644 packages/.github/prompts/opsx-apply.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-archive.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-explore.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-propose.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-sync.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-update.prompt.md
 delete mode 100644 packages/.github/prompts/opsx-verify.prompt.md
 delete mode 100644 packages/.github/skills/openspec-apply-change/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-archive-change/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-explore/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-propose/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-sync-specs/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-update-change/SKILL.md
 delete mode 100644 packages/.github/skills/openspec-verify-change/SKILL.md
 delete mode 100644 packages/openspec/config.yaml

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
new file mode 100644
index 00000000..4f37d580
--- /dev/null
+++ b/CONTRIBUTING.md
@@ -0,0 +1,142 @@
+# Contributing to @editorjs/document-model
+
+Thanks for your interest in contributing! This repo is a Yarn 4 (Berry) workspaces monorepo containing the model-driven, collaboration-ready Editor.js engine. This guide covers local setup, day-to-day development, and — since this project practices **spec-driven development with [OpenSpec](https://github.com/Fission-AI/OpenSpec)** — how specs and change proposals fit into the workflow.
+
+For a general, non-technical overview of ways to get involved, see the [Want to contribute?](README.md#want-to-contribute) section of the README and [editorjs.io/contributing](https://editorjs.io/contributing/).
+
+## Getting started
+
+**Prerequisites**
+- Node.js `v24.15.0` (see [`.nvmrc`](.nvmrc) — use `nvm use` if you manage versions with nvm)
+- Yarn `4.0.1` via [Corepack](https://yarnpkg.com/corepack) (`corepack enable`)
+
+**Setup**
+```bash
+yarn install
+```
+
+This installs dependencies for every workspace under `packages/*` and `packages/tools/*`, and sets up the Husky git hook (`yarn constraints` runs on pre-commit to keep workspace dependency versions consistent).
+
+**Common commands** (from the repo root, across all packages):
+```bash
+yarn build        # build every package
+yarn test          # run every package's test suite
+yarn lint          # lint every package
+yarn lint:fix       # lint and auto-fix every package
+```
+
+To work within a single package:
+```bash
+cd packages/model
+yarn test
+yarn test:coverage
+yarn lint
+```
+
+Some packages (e.g. `model`) also run Stryker mutation testing via `yarn test:mutations`.
+
+To try changes end-to-end, use the Vite playground:
+```bash
+cd packages/playground && yarn dev
+```
+
+The OT collaboration server can be run via Docker: `docker compose up`.
+
+## Project structure
+
+- `packages/model-types` — shared low-level types/event base classes (internal to `model`/`sdk`)
+- `packages/sdk` — public contracts (`EventBus`, interfaces) that tools/plugins should depend on
+- `packages/model` — the in-memory document model (`EditorJSModel`, `BlockNode`, `TextNode`, caret management)
+- `packages/dom-adapters` — binds model nodes to DOM inputs
+- `packages/collaboration-manager` — OT, batching, undo/redo, OT WebSocket client
+- `packages/core` — orchestrator (IoC container, plugin/tool lifecycle, `EditorAPI`)
+- `packages/ui` — default UI shell
+- `packages/ot-server` — standalone WebSocket OT server
+- `packages/playground` — manual testing sandbox
+- `packages/tools/*` — built-in block/inline tools (paragraph, bold, italic, inline-link), each implementing `BlockTool`/`InlineTool` from `@editorjs/sdk`
+
+New tools/plugins should depend on `@editorjs/sdk`, not `@editorjs/model` directly.
+
+Architecture, data model, input handling, plugin lifecycle, collaboration, and event system docs live in [`docs/`](docs/README.md), with Mermaid diagrams in `docs/diagrams/`. Read the relevant page before making structural changes — and expect PRs that touch public APIs to require doc updates (see [Documentation](#documentation) below).
+
+## Spec-driven development with OpenSpec
+
+This repo's source-of-truth specs live in [`openspec/specs/`](openspec/specs) (one folder per package/area, e.g. `model`, `core`, `collaboration-manager`, `sdk`, `architecture`). Project-wide conventions for AI-assisted work are configured in [`openspec/config.yaml`](openspec/config.yaml), including:
+
+- **TDD is required**: write a failing test first, then implement, then refactor.
+- **Test naming**: use should-notation, e.g. `it('should return X when Y')`.
+- **Proposals must cross-check `docs/`** and call out anything they supersede.
+
+### When to write a change proposal
+
+For any non-trivial change (new capability, breaking change, architectural shift, or anything that alters behavior described in `openspec/specs/`), start with an OpenSpec change proposal instead of jumping straight into code. Small fixes, typo corrections, and refactors that don't change behavior don't need one — use your judgment, and feel free to ask in a draft PR or issue if unsure.
+
+### Workflow
+
+The [OpenSpec CLI](https://github.com/Fission-AI/OpenSpec) (`@fission-ai/openspec`) drives this. Install it globally or use `npx openspec`:
+
+```bash
+npm install -g @fission-ai/openspec
+```
+
+Typical flow for a change:
+
+1. **Explore** (optional) — think through the problem and requirements before committing to an approach.
+2. **Propose** — create a change under `openspec/changes/<change-name>/` with `proposal.md` (what & why), `design.md` (how), and `tasks.md` (implementation steps):
+   ```bash
+   openspec new change "<change-name>"
+   openspec status --change "<change-name>"
+   ```
+3. **Apply** — implement the tasks in `tasks.md`, following TDD as required by `openspec/config.yaml`.
+4. **Verify** — check the implementation matches the proposal's specs and tasks before wrapping up.
+5. **Archive / Sync** — once merged, archive the change (`openspec archive <change-name>`) to fold its delta specs into `openspec/specs/`, or sync specs without archiving if the change is still in flight.
+
+Useful CLI commands while working on a change:
+```bash
+openspec list                       # list active changes/specs
+openspec show <item-name>           # show a change or spec
+openspec validate <item-name>       # validate a change/spec against its schema
+openspec view                       # interactive dashboard
+```
+
+If you're using Claude Code in this repo, the same workflow is available as slash commands (`/opsx:explore`, `/opsx:propose`, `/opsx:apply`, `/opsx:verify`, `/opsx:sync`, `/opsx:archive`, `/opsx:update`) and equivalent skills — these wrap the CLI above and keep proposal/spec/task files coherent as you iterate.
+
+Keep specs and code in sync: a PR that changes behavior covered by `openspec/specs/` without a corresponding proposal/archive will likely be asked to add one during review.
+
+## Testing
+
+- Tests are co-located with source as `*.spec.ts` (unit) and `*.integration.spec.ts` (integration), run with Jest.
+- Name test cases in should-notation: `it('should return X when Y')`.
+- Write the failing test before the implementation (TDD), per `openspec/config.yaml`.
+- Some packages run Stryker mutation testing (`yarn test:mutations`); CI runs mutation tests on changed files for PRs and on all files for merge groups.
+
+## Linting and code style
+
+- ESLint (`eslint-config-codex`) is the source of truth for style; run `yarn lint` (or `yarn lint:fix`) before pushing.
+- CI enforces `lint:ci` with zero warnings allowed per package.
+
+## Commit messages
+
+Recent history favors [Conventional Commits](https://www.conventionalcommits.org/)-style messages scoped to a package, e.g.:
+```
+fix(sdk): remove @/ alias from tsconfig
+refactor(link): extract link inline tool into standalone package
+```
+Use a scope matching the package you changed where it makes sense (`model`, `core`, `sdk`, `collaboration-manager`, etc.).
+
+## Documentation
+
+Public API or architectural changes should be reflected in `docs/` (see the map in [`.github/agents/docs-updater.agent.md`](.github/agents/docs-updater.agent.md) for which doc/diagram pairs correspond to which parts of the codebase). Diagrams are Mermaid files in `docs/diagrams/` — keep node/class names in sync with actual TypeScript identifiers, and never invent methods or properties that don't exist.
+
+## Opening a pull request
+
+- CI runs per affected package (lint, unit tests, build, and mutation tests where configured) — only packages whose files (or dependencies) changed will run.
+- Make sure `yarn build`, `yarn test`, and `yarn lint` pass locally for any package you touched.
+- If your change is behavior-affecting, link the relevant OpenSpec change (or note that one wasn't needed and why).
+- Keep PRs scoped to one package/concern where possible, matching the existing commit convention.
+
+## Getting help
+
+- [Good First Tasks](https://github.com/codex-team/editor.js/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+task%22)
+- [Telegram Chat](https://t.me/codex_editor)
+- Custom requirements: team@codex.so
diff --git a/packages/.claude/commands/opsx/apply.md b/packages/.claude/commands/opsx/apply.md
deleted file mode 100644
index c6cb9b6d..00000000
--- a/packages/.claude/commands/opsx/apply.md
+++ /dev/null
@@ -1,156 +0,0 @@
----
-name: "OPSX: Apply"
-description: Implement tasks from an OpenSpec change (Experimental)
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, artifacts, experimental]
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! You can archive this change with `/opsx:archive`.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.claude/commands/opsx/archive.md b/packages/.claude/commands/opsx/archive.md
deleted file mode 100644
index df8a2f2d..00000000
--- a/packages/.claude/commands/opsx/archive.md
+++ /dev/null
@@ -1,161 +0,0 @@
----
-name: "OPSX: Archive"
-description: Archive a completed change in the experimental workflow
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, archive, experimental]
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Spec sync status (synced / sync skipped / no delta specs)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success (No Delta Specs)**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** No delta specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success With Warnings**
-
-```
-## Archive Complete (with warnings)
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** Sync skipped (user chose to skip)
-
-**Warnings:**
-- Archived with 2 incomplete artifacts
-- Archived with 3 incomplete tasks
-- Delta spec sync was skipped (user chose to skip)
-
-Review the archive if this was not intentional.
-```
-
-**Output On Error (Archive Exists)**
-
-```
-## Archive Failed
-
-**Change:** <change-name>
-**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-
-Target archive directory already exists.
-
-**Options:**
-1. Rename the existing archive
-2. Delete the existing archive if it's a duplicate
-3. Wait until a different date to archive
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.claude/commands/opsx/explore.md b/packages/.claude/commands/opsx/explore.md
deleted file mode 100644
index 7558a631..00000000
--- a/packages/.claude/commands/opsx/explore.md
+++ /dev/null
@@ -1,175 +0,0 @@
----
-name: "OPSX: Explore"
-description: "Enter explore mode - think through ideas, investigate problems, clarify requirements"
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, explore, experimental, thinking]
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
-- A vague idea: "real-time collaboration"
-- A specific problem: "the auth system is getting unwieldy"
-- A change name: "add-dark-mode" (to explore in context of that change)
-- A comparison: "postgres vs sqlite for this"
-- Nothing (just enter explore mode)
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-If the user mentioned a specific change name, read its artifacts for context.
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.claude/commands/opsx/propose.md b/packages/.claude/commands/opsx/propose.md
deleted file mode 100644
index 8d99588f..00000000
--- a/packages/.claude/commands/opsx/propose.md
+++ /dev/null
@@ -1,110 +0,0 @@
----
-name: "OPSX: Propose"
-description: Propose a new change - create it and generate all artifacts in one step
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, artifacts, experimental]
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
-
-**Steps**
-
-1. **If no input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` to start implementing."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.claude/commands/opsx/sync.md b/packages/.claude/commands/opsx/sync.md
deleted file mode 100644
index a3550cf4..00000000
--- a/packages/.claude/commands/opsx/sync.md
+++ /dev/null
@@ -1,144 +0,0 @@
----
-name: "OPSX: Sync"
-description: Sync delta specs from a change to main specs
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, specs, experimental]
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/packages/.claude/commands/opsx/update.md b/packages/.claude/commands/opsx/update.md
deleted file mode 100644
index 75104f59..00000000
--- a/packages/.claude/commands/opsx/update.md
+++ /dev/null
@@ -1,82 +0,0 @@
----
-name: "OPSX: Update"
-description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, artifacts, experimental]
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.claude/commands/opsx/verify.md b/packages/.claude/commands/opsx/verify.md
deleted file mode 100644
index 7f6cade1..00000000
--- a/packages/.claude/commands/opsx/verify.md
+++ /dev/null
@@ -1,168 +0,0 @@
----
-name: "OPSX: Verify"
-description: Verify implementation matches change artifacts before archiving
-allowed-tools: Bash(openspec:*)
-category: Workflow
-tags: [workflow, verify, experimental]
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/packages/.claude/skills/openspec-apply-change/SKILL.md b/packages/.claude/skills/openspec-apply-change/SKILL.md
deleted file mode 100644
index 1375861c..00000000
--- a/packages/.claude/skills/openspec-apply-change/SKILL.md
+++ /dev/null
@@ -1,160 +0,0 @@
----
-name: openspec-apply-change
-description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! Ready to archive this change.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.claude/skills/openspec-archive-change/SKILL.md b/packages/.claude/skills/openspec-archive-change/SKILL.md
deleted file mode 100644
index c0c169d1..00000000
--- a/packages/.claude/skills/openspec-archive-change/SKILL.md
+++ /dev/null
@@ -1,118 +0,0 @@
----
-name: openspec-archive-change
-description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Whether specs were synced (if applicable)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
-
-All artifacts complete. All tasks complete.
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use openspec-sync-specs approach (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.claude/skills/openspec-explore/SKILL.md b/packages/.claude/skills/openspec-explore/SKILL.md
deleted file mode 100644
index 771271a8..00000000
--- a/packages/.claude/skills/openspec-explore/SKILL.md
+++ /dev/null
@@ -1,290 +0,0 @@
----
-name: openspec-explore
-description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Handling Different Entry Points
-
-**User brings a vague idea:**
-```
-User: I'm thinking about adding real-time collaboration
-
-You: Real-time collab is a big space. Let me think about this...
-
-      COLLABORATION SPECTRUM
-      ════════════════════════════════════════════
-
-      Awareness          Coordination         Sync
-          │                   │                 │
-          ▼                   ▼                 ▼
-      ┌────────┐         ┌────────┐        ┌────────┐
-      │Presence│         │Cursors │        │  CRDT  │
-      │  "3    │         │ Multi  │        │Conflict│
-      │online" │         │ select │        │  free  │
-      └────────┘         └────────┘        └────────┘
-          │                   │                 │
-       trivial            moderate           complex
-
-      Where's your head at?
-```
-
-**User brings a specific problem:**
-```
-User: The auth system is a mess
-
-You: [reads codebase]
-
-     ┌─────────────────────────────────────────────┐
-     │              CURRENT AUTH FLOW              │
-     └─────────────────────────────────────────────┘
-                        │
-          ┌─────────────┼─────────────┐
-          ▼             ▼             ▼
-     ┌─────────┐  ┌─────────┐  ┌─────────┐
-     │ Google  │  │  GitHub │  │  Email  │
-     │  OAuth  │  │  OAuth  │  │  Magic  │
-     └────┬────┘  └────┬────┘  └────┬────┘
-          │             │             │
-          └─────────────┼─────────────┘
-                        ▼
-                  ┌───────────┐
-                  │  Session  │
-                  └─────┬─────┘
-                        │
-                        ▼
-                  ┌───────────┐
-                  │   Perms   │
-                  └───────────┘
-
-     I see three tangles. Which one's burning?
-```
-
-**User is stuck mid-implementation:**
-```
-User: /opsx:explore add-auth-system
-      The OAuth integration is more complex than expected
-
-You: [reads change artifacts]
-
-     You're on task 4: "Implement OAuth flow"
-
-     Let me trace what's involved...
-
-     [draws diagram, explores options, suggests paths]
-
-     Want to update the design to reflect this?
-     Or add a spike task to investigate?
-```
-
-**User wants to compare options:**
-```
-User: Should we use Postgres or SQLite?
-
-You: Generic answer is boring. What's the context?
-
-User: A CLI tool that tracks local dev environments
-
-You: That changes everything.
-
-     ┌─────────────────────────────────────────────────┐
-     │          CLI TOOL DATA STORAGE                  │
-     └─────────────────────────────────────────────────┘
-
-     Key constraints:
-     • No daemon running
-     • Must work offline
-     • Single user
-
-                  SQLite          Postgres
-     Deployment   embedded ✓      needs server ✗
-     Offline      yes ✓           no ✗
-     Single file  yes ✓           no ✗
-
-     SQLite. Not even close.
-
-     Unless... is there a sync component?
-```
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When it feels like things are crystallizing, you might summarize:
-
-```
-## What We Figured Out
-
-**The problem**: [crystallized understanding]
-
-**The approach**: [if one emerged]
-
-**Open questions**: [if any remain]
-
-**Next steps** (if ready):
-- Create a change proposal
-- Keep exploring: just keep talking
-```
-
-But this summary is optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.claude/skills/openspec-propose/SKILL.md b/packages/.claude/skills/openspec-propose/SKILL.md
deleted file mode 100644
index 716d2d3d..00000000
--- a/packages/.claude/skills/openspec-propose/SKILL.md
+++ /dev/null
@@ -1,114 +0,0 @@
----
-name: openspec-propose
-description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
-
-**Steps**
-
-1. **If no clear input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.claude/skills/openspec-sync-specs/SKILL.md b/packages/.claude/skills/openspec-sync-specs/SKILL.md
deleted file mode 100644
index b6aad67a..00000000
--- a/packages/.claude/skills/openspec-sync-specs/SKILL.md
+++ /dev/null
@@ -1,148 +0,0 @@
----
-name: openspec-sync-specs
-description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/packages/.claude/skills/openspec-update-change/SKILL.md b/packages/.claude/skills/openspec-update-change/SKILL.md
deleted file mode 100644
index 3e1c931d..00000000
--- a/packages/.claude/skills/openspec-update-change/SKILL.md
+++ /dev/null
@@ -1,86 +0,0 @@
----
-name: openspec-update-change
-description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.claude/skills/openspec-verify-change/SKILL.md b/packages/.claude/skills/openspec-verify-change/SKILL.md
deleted file mode 100644
index b6557c6f..00000000
--- a/packages/.claude/skills/openspec-verify-change/SKILL.md
+++ /dev/null
@@ -1,172 +0,0 @@
----
-name: openspec-verify-change
-description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/packages/.github/prompts/opsx-apply.prompt.md b/packages/.github/prompts/opsx-apply.prompt.md
deleted file mode 100644
index 5bd0b6bb..00000000
--- a/packages/.github/prompts/opsx-apply.prompt.md
+++ /dev/null
@@ -1,152 +0,0 @@
----
-description: Implement tasks from an OpenSpec change (Experimental)
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name (e.g., `/opsx:apply add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using `/opsx:continue`
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! You can archive this change with `/opsx:archive`.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.github/prompts/opsx-archive.prompt.md b/packages/.github/prompts/opsx-archive.prompt.md
deleted file mode 100644
index f9ba89d8..00000000
--- a/packages/.github/prompts/opsx-archive.prompt.md
+++ /dev/null
@@ -1,157 +0,0 @@
----
-description: Archive a completed change in the experimental workflow
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:archive` (e.g., `/opsx:archive add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Prompt user for confirmation to continue
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Spec sync status (synced / sync skipped / no delta specs)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success (No Delta Specs)**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** No delta specs
-
-All artifacts complete. All tasks complete.
-```
-
-**Output On Success With Warnings**
-
-```
-## Archive Complete (with warnings)
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** Sync skipped (user chose to skip)
-
-**Warnings:**
-- Archived with 2 incomplete artifacts
-- Archived with 3 incomplete tasks
-- Delta spec sync was skipped (user chose to skip)
-
-Review the archive if this was not intentional.
-```
-
-**Output On Error (Archive Exists)**
-
-```
-## Archive Failed
-
-**Change:** <change-name>
-**Target:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-
-Target archive directory already exists.
-
-**Options:**
-1. Rename the existing archive
-2. Delete the existing archive if it's a duplicate
-3. Wait until a different date to archive
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use the Skill tool to invoke `openspec-sync-specs` (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.github/prompts/opsx-explore.prompt.md b/packages/.github/prompts/opsx-explore.prompt.md
deleted file mode 100644
index 856001a6..00000000
--- a/packages/.github/prompts/opsx-explore.prompt.md
+++ /dev/null
@@ -1,171 +0,0 @@
----
-description: Enter explore mode - think through ideas, investigate problems, clarify requirements
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:explore` is whatever the user wants to think about. Could be:
-- A vague idea: "real-time collaboration"
-- A specific problem: "the auth system is getting unwieldy"
-- A change name: "add-dark-mode" (to explore in context of that change)
-- A comparison: "postgres vs sqlite for this"
-- Nothing (just enter explore mode)
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-If the user mentioned a specific change name, read its artifacts for context.
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When things crystallize, you might offer a summary - but it's optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.github/prompts/opsx-propose.prompt.md b/packages/.github/prompts/opsx-propose.prompt.md
deleted file mode 100644
index fde349a6..00000000
--- a/packages/.github/prompts/opsx-propose.prompt.md
+++ /dev/null
@@ -1,106 +0,0 @@
----
-description: Propose a new change - create it and generate all artifacts in one step
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The argument after `/opsx:propose` is the change name (kebab-case), OR a description of what the user wants to build.
-
-**Steps**
-
-1. **If no input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` to start implementing."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.github/prompts/opsx-sync.prompt.md b/packages/.github/prompts/opsx-sync.prompt.md
deleted file mode 100644
index 961aa409..00000000
--- a/packages/.github/prompts/opsx-sync.prompt.md
+++ /dev/null
@@ -1,140 +0,0 @@
----
-description: Sync delta specs from a change to main specs
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:sync` (e.g., `/opsx:sync add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/packages/.github/prompts/opsx-update.prompt.md b/packages/.github/prompts/opsx-update.prompt.md
deleted file mode 100644
index 5b316d4d..00000000
--- a/packages/.github/prompts/opsx-update.prompt.md
+++ /dev/null
@@ -1,78 +0,0 @@
----
-description: Update a change - revise existing planning artifacts and keep them coherent (Experimental)
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:update` (e.g., `/opsx:update add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.github/prompts/opsx-verify.prompt.md b/packages/.github/prompts/opsx-verify.prompt.md
deleted file mode 100644
index cc1377a7..00000000
--- a/packages/.github/prompts/opsx-verify.prompt.md
+++ /dev/null
@@ -1,164 +0,0 @@
----
-description: Verify implementation matches change artifacts before archiving
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name after `/opsx:verify` (e.g., `/opsx:verify add-auth`). If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/packages/.github/skills/openspec-apply-change/SKILL.md b/packages/.github/skills/openspec-apply-change/SKILL.md
deleted file mode 100644
index 1375861c..00000000
--- a/packages/.github/skills/openspec-apply-change/SKILL.md
+++ /dev/null
@@ -1,160 +0,0 @@
----
-name: openspec-apply-change
-description: Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Implement tasks from an OpenSpec change.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **Select the change**
-
-   If a name is provided, use it. Otherwise:
-   - Infer from conversation context if the user mentioned a change
-   - Auto-select if only one active change exists
-   - If ambiguous, run `openspec list --json` to get available changes and use the **AskUserQuestion tool** to let the user select
-
-   Always announce: "Using change: <name>" and how to override (e.g., `/opsx:apply <other>`).
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, and `actionContext`: planning scope and edit constraints
-   - Which artifact contains the tasks (typically "tasks" for spec-driven, check status for others)
-
-3. **Get apply instructions**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns:
-   - `contextFiles`: artifact ID -> array of concrete file paths (varies by schema - could be proposal/specs/design/tasks or spec/tests/implementation/docs)
-   - Progress (total, complete, remaining)
-   - Task list with status
-   - Dynamic instruction based on current state
-
-   **Handle states:**
-   - If `state: "blocked"` (missing artifacts): show message, suggest using openspec-continue-change
-   - If `state: "all_done"`: congratulate, suggest archive
-   - Otherwise: proceed to implementation
-
-4. **Read context files**
-
-   Read every file path listed under `contextFiles` from the apply instructions output.
-   The files depend on the schema being used:
-   - **spec-driven**: proposal, specs, design, tasks
-   - Other schemas: follow the contextFiles from CLI output
-
-5. **Show current progress**
-
-   Display:
-   - Schema being used
-   - Progress: "N/M tasks complete"
-   - Remaining tasks overview
-   - Dynamic instruction from CLI
-
-6. **Implement tasks (loop until done or blocked)**
-
-   For each pending task:
-   - Show which task is being worked on
-   - Make the code changes required
-   - Keep changes minimal and focused
-   - Mark task complete in the tasks file: `- [ ]` → `- [x]`
-   - Continue to next task
-
-   **Pause if:**
-   - Task is unclear → ask for clarification
-   - Implementation reveals a design issue → suggest updating artifacts
-   - Error or blocker encountered → report and wait for guidance
-   - User interrupts
-
-7. **On completion or pause, show status**
-
-   Display:
-   - Tasks completed this session
-   - Overall progress: "N/M tasks complete"
-   - If all done: suggest archive
-   - If paused: explain why and wait for guidance
-
-**Output During Implementation**
-
-```
-## Implementing: <change-name> (schema: <schema-name>)
-
-Working on task 3/7: <task description>
-[...implementation happening...]
-✓ Task complete
-
-Working on task 4/7: <task description>
-[...implementation happening...]
-✓ Task complete
-```
-
-**Output On Completion**
-
-```
-## Implementation Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 7/7 tasks complete ✓
-
-### Completed This Session
-- [x] Task 1
-- [x] Task 2
-...
-
-All tasks complete! Ready to archive this change.
-```
-
-**Output On Pause (Issue Encountered)**
-
-```
-## Implementation Paused
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Progress:** 4/7 tasks complete
-
-### Issue Encountered
-<description of the issue>
-
-**Options:**
-1. <option 1>
-2. <option 2>
-3. Other approach
-
-What would you like to do?
-```
-
-**Guardrails**
-- Keep going through tasks until done or blocked
-- Always read context files before starting (from the apply instructions output)
-- If task is ambiguous, pause and ask before implementing
-- If implementation reveals issues, pause and suggest artifact updates
-- Keep code changes minimal and scoped to each task
-- Update task checkbox immediately after completing each task
-- Pause on errors, blockers, or unclear requirements - don't guess
-- Use contextFiles from CLI output, don't assume specific file names
-
-**Fluid Workflow Integration**
-
-This skill supports the "actions on a change" model:
-
-- **Can be invoked anytime**: Before all artifacts are done (if tasks exist), after partial implementation, interleaved with other actions
-- **Allows artifact updates**: If implementation reveals design issues, suggest updating artifacts - not phase-locked, work fluidly
diff --git a/packages/.github/skills/openspec-archive-change/SKILL.md b/packages/.github/skills/openspec-archive-change/SKILL.md
deleted file mode 100644
index c0c169d1..00000000
--- a/packages/.github/skills/openspec-archive-change/SKILL.md
+++ /dev/null
@@ -1,118 +0,0 @@
----
-name: openspec-archive-change
-description: Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Archive a completed change in the experimental workflow.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show only active changes (not already archived).
-   Include the schema used for each change if available.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check artifact completion status**
-
-   Run `openspec status --change "<name>" --json` to check artifact completion.
-
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - `artifacts`: List of artifacts with their status (`done` or other)
-
-   **If any artifacts are not `done`:**
-   - Display warning listing incomplete artifacts
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-3. **Check task completion status**
-
-   Read the tasks file (typically `tasks.md`) to check for incomplete tasks.
-
-   Count tasks marked with `- [ ]` (incomplete) vs `- [x]` (complete).
-
-   **If incomplete tasks found:**
-   - Display warning showing count of incomplete tasks
-   - Use **AskUserQuestion tool** to confirm user wants to proceed
-   - Proceed if user confirms
-
-   **If no tasks file exists:** Proceed without task-related warning.
-
-4. **Assess delta spec sync state**
-
-   Use `artifactPaths.specs.existingOutputPaths` from status JSON to check for delta specs. If none exist, proceed without sync prompt.
-
-   **If delta specs exist:**
-   - Compare each delta spec with its corresponding main spec at `openspec/specs/<capability>/spec.md`
-   - Determine what changes would be applied (adds, modifications, removals, renames)
-   - Show a combined summary before prompting
-
-   **Prompt options:**
-   - If changes needed: "Sync now (recommended)", "Archive without syncing"
-   - If already synced: "Archive now", "Sync anyway", "Cancel"
-
-   If user chooses sync, use Task tool (subagent_type: "general-purpose", prompt: "Use Skill tool to invoke openspec-sync-specs for change '<name>'. Delta spec analysis: <include the analyzed delta spec summary>"). Proceed to archive regardless of choice.
-
-5. **Perform the archive**
-
-   Create an `archive` directory under `planningHome.changesDir` if it doesn't exist:
-   ```bash
-   mkdir -p "<planningHome.changesDir>/archive"
-   ```
-
-   Generate target name using current date: `YYYY-MM-DD-<change-name>`
-
-   **Check if target already exists:**
-   - If yes: Fail with error, suggest renaming existing archive or using different date
-   - If no: Move `changeRoot` to the archive directory
-
-   ```bash
-   mv "<changeRoot>" "<planningHome.changesDir>/archive/YYYY-MM-DD-<name>"
-   ```
-
-6. **Display summary**
-
-   Show archive completion summary including:
-   - Change name
-   - Schema that was used
-   - Archive location
-   - Whether specs were synced (if applicable)
-   - Note about any warnings (incomplete artifacts/tasks)
-
-**Output On Success**
-
-```
-## Archive Complete
-
-**Change:** <change-name>
-**Schema:** <schema-name>
-**Archived to:** the archive path derived from `planningHome.changesDir`/YYYY-MM-DD-<name>/
-**Specs:** ✓ Synced to main specs (or "No delta specs" or "Sync skipped")
-
-All artifacts complete. All tasks complete.
-```
-
-**Guardrails**
-- Always prompt for change selection if not provided
-- Use artifact graph (openspec status --json) for completion checking
-- Don't block archive on warnings - just inform and confirm
-- Preserve .openspec.yaml when moving to archive (it moves with the directory)
-- Show clear summary of what happened
-- If sync is requested, use openspec-sync-specs approach (agent-driven)
-- If delta specs exist, always run the sync assessment and show the combined summary before prompting
diff --git a/packages/.github/skills/openspec-explore/SKILL.md b/packages/.github/skills/openspec-explore/SKILL.md
deleted file mode 100644
index 771271a8..00000000
--- a/packages/.github/skills/openspec-explore/SKILL.md
+++ /dev/null
@@ -1,290 +0,0 @@
----
-name: openspec-explore
-description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
-
-**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER write code or implement features. If the user asks you to implement something, remind them to exit explore mode first and create a change proposal. You MAY create OpenSpec artifacts (proposals, designs, specs) if the user asks—that's capturing thinking, not implementing.
-
-**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
----
-
-## The Stance
-
-- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
-- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
-- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
-- **Adaptive** - Follow interesting threads, pivot when new information emerges
-- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
-- **Grounded** - Explore the actual codebase when relevant, don't just theorize
-
----
-
-## What You Might Do
-
-Depending on what the user brings, you might:
-
-**Explore the problem space**
-- Ask clarifying questions that emerge from what they said
-- Challenge assumptions
-- Reframe the problem
-- Find analogies
-
-**Investigate the codebase**
-- Map existing architecture relevant to the discussion
-- Find integration points
-- Identify patterns already in use
-- Surface hidden complexity
-
-**Compare options**
-- Brainstorm multiple approaches
-- Build comparison tables
-- Sketch tradeoffs
-- Recommend a path (if asked)
-
-**Visualize**
-```
-┌─────────────────────────────────────────┐
-│     Use ASCII diagrams liberally        │
-├─────────────────────────────────────────┤
-│                                         │
-│      ┌────────┐         ┌────────┐      │
-│      │ State  │────────▶│ State  │      │
-│      │   A    │         │   B    │      │
-│      └────────┘         └────────┘      │
-│                                         │
-│   System diagrams, state machines,      │
-│   data flows, architecture sketches,    │
-│   dependency graphs, comparison tables  │
-│                                         │
-└─────────────────────────────────────────┘
-```
-
-**Surface risks and unknowns**
-- Identify what could go wrong
-- Find gaps in understanding
-- Suggest spikes or investigations
-
----
-
-## OpenSpec Awareness
-
-You have full context of the OpenSpec system. Use it naturally, don't force it.
-
-### Check for context
-
-At the start, quickly check what exists:
-```bash
-openspec list --json
-```
-
-This tells you:
-- If there are active changes
-- Their names, schemas, and status
-- What the user might be working on
-
-### When no change exists
-
-Think freely. When insights crystallize, you might offer:
-
-- "This feels solid enough to start a change. Want me to create a proposal?"
-- Or keep exploring - no pressure to formalize
-
-### When a change exists
-
-If the user mentions a change or you detect one is relevant:
-
-1. **Resolve and read existing artifacts for context**
-   - Run `openspec status --change "<name>" --json`.
-   - Use `changeRoot`, `artifactPaths`, and `actionContext` from the status JSON.
-   - Read existing files from `artifactPaths.<artifact>.existingOutputPaths`.
-
-2. **Reference them naturally in conversation**
-   - "Your design mentions using Redis, but we just realized SQLite fits better..."
-   - "The proposal scopes this to premium users, but we're now thinking everyone..."
-
-3. **Offer to capture when decisions are made**
-
-    | Insight Type               | Where to Capture               |
-    |----------------------------|--------------------------------|
-    | New requirement discovered | `specs/<capability>/spec.md` |
-    | Requirement changed        | `specs/<capability>/spec.md` |
-    | Design decision made       | `design.md`                  |
-    | Scope changed              | `proposal.md`                |
-    | New work identified        | `tasks.md`                   |
-    | Assumption invalidated     | Relevant artifact              |
-
-   Example offers:
-   - "That's a design decision. Capture it in design.md?"
-   - "This is a new requirement. Add it to specs?"
-   - "This changes scope. Update the proposal?"
-
-4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
-
----
-
-## What You Don't Have To Do
-
-- Follow a script
-- Ask the same questions every time
-- Produce a specific artifact
-- Reach a conclusion
-- Stay on topic if a tangent is valuable
-- Be brief (this is thinking time)
-
----
-
-## Handling Different Entry Points
-
-**User brings a vague idea:**
-```
-User: I'm thinking about adding real-time collaboration
-
-You: Real-time collab is a big space. Let me think about this...
-
-      COLLABORATION SPECTRUM
-      ════════════════════════════════════════════
-
-      Awareness          Coordination         Sync
-          │                   │                 │
-          ▼                   ▼                 ▼
-      ┌────────┐         ┌────────┐        ┌────────┐
-      │Presence│         │Cursors │        │  CRDT  │
-      │  "3    │         │ Multi  │        │Conflict│
-      │online" │         │ select │        │  free  │
-      └────────┘         └────────┘        └────────┘
-          │                   │                 │
-       trivial            moderate           complex
-
-      Where's your head at?
-```
-
-**User brings a specific problem:**
-```
-User: The auth system is a mess
-
-You: [reads codebase]
-
-     ┌─────────────────────────────────────────────┐
-     │              CURRENT AUTH FLOW              │
-     └─────────────────────────────────────────────┘
-                        │
-          ┌─────────────┼─────────────┐
-          ▼             ▼             ▼
-     ┌─────────┐  ┌─────────┐  ┌─────────┐
-     │ Google  │  │  GitHub │  │  Email  │
-     │  OAuth  │  │  OAuth  │  │  Magic  │
-     └────┬────┘  └────┬────┘  └────┬────┘
-          │             │             │
-          └─────────────┼─────────────┘
-                        ▼
-                  ┌───────────┐
-                  │  Session  │
-                  └─────┬─────┘
-                        │
-                        ▼
-                  ┌───────────┐
-                  │   Perms   │
-                  └───────────┘
-
-     I see three tangles. Which one's burning?
-```
-
-**User is stuck mid-implementation:**
-```
-User: /opsx:explore add-auth-system
-      The OAuth integration is more complex than expected
-
-You: [reads change artifacts]
-
-     You're on task 4: "Implement OAuth flow"
-
-     Let me trace what's involved...
-
-     [draws diagram, explores options, suggests paths]
-
-     Want to update the design to reflect this?
-     Or add a spike task to investigate?
-```
-
-**User wants to compare options:**
-```
-User: Should we use Postgres or SQLite?
-
-You: Generic answer is boring. What's the context?
-
-User: A CLI tool that tracks local dev environments
-
-You: That changes everything.
-
-     ┌─────────────────────────────────────────────────┐
-     │          CLI TOOL DATA STORAGE                  │
-     └─────────────────────────────────────────────────┘
-
-     Key constraints:
-     • No daemon running
-     • Must work offline
-     • Single user
-
-                  SQLite          Postgres
-     Deployment   embedded ✓      needs server ✗
-     Offline      yes ✓           no ✗
-     Single file  yes ✓           no ✗
-
-     SQLite. Not even close.
-
-     Unless... is there a sync component?
-```
-
----
-
-## Ending Discovery
-
-There's no required ending. Discovery might:
-
-- **Flow into a proposal**: "Ready to start? I can create a change proposal."
-- **Result in artifact updates**: "Updated design.md with these decisions"
-- **Just provide clarity**: User has what they need, moves on
-- **Continue later**: "We can pick this up anytime"
-
-When it feels like things are crystallizing, you might summarize:
-
-```
-## What We Figured Out
-
-**The problem**: [crystallized understanding]
-
-**The approach**: [if one emerged]
-
-**Open questions**: [if any remain]
-
-**Next steps** (if ready):
-- Create a change proposal
-- Keep exploring: just keep talking
-```
-
-But this summary is optional. Sometimes the thinking IS the value.
-
----
-
-## Guardrails
-
-- **Don't implement** - Never write code or implement features. Creating OpenSpec artifacts is fine, writing application code is not.
-- **Don't fake understanding** - If something is unclear, dig deeper
-- **Don't rush** - Discovery is thinking time, not task time
-- **Don't force structure** - Let patterns emerge naturally
-- **Don't auto-capture** - Offer to save insights, don't just do it
-- **Do visualize** - A good diagram is worth many paragraphs
-- **Do explore the codebase** - Ground discussions in reality
-- **Do question assumptions** - Including the user's and your own
diff --git a/packages/.github/skills/openspec-propose/SKILL.md b/packages/.github/skills/openspec-propose/SKILL.md
deleted file mode 100644
index 716d2d3d..00000000
--- a/packages/.github/skills/openspec-propose/SKILL.md
+++ /dev/null
@@ -1,114 +0,0 @@
----
-name: openspec-propose
-description: Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Propose a new change - create the change and generate all artifacts in one step.
-
-I'll create a change with artifacts:
-- proposal.md (what & why)
-- design.md (how)
-- tasks.md (implementation steps)
-
-When ready to implement, run /opsx:apply
-
----
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: The user's request should include a change name (kebab-case) OR a description of what they want to build.
-
-**Steps**
-
-1. **If no clear input provided, ask what they want to build**
-
-   Use the **AskUserQuestion tool** (open-ended, no preset options) to ask:
-   > "What change do you want to work on? Describe what you want to build or fix."
-
-   From their description, derive a kebab-case name (e.g., "add user authentication" → `add-user-auth`).
-
-   **IMPORTANT**: Do NOT proceed without understanding what the user wants to build.
-
-2. **Create the change directory**
-   ```bash
-   openspec new change "<name>"
-   ```
-   This creates a scaffolded change in the planning home resolved by the CLI with `.openspec.yaml`.
-
-3. **Get the artifact build order**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to get:
-   - `applyRequires`: array of artifact IDs needed before implementation (e.g., `["tasks"]`)
-   - `artifacts`: list of all artifacts with their status and dependencies
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-4. **Create artifacts in sequence until apply-ready**
-
-   Use the **TodoWrite tool** to track progress through the artifacts.
-
-   Loop through artifacts in dependency order (artifacts with no pending dependencies first):
-
-   a. **For each artifact that is `ready` (dependencies satisfied)**:
-      - Get instructions:
-        ```bash
-        openspec instructions <artifact-id> --change "<name>" --json
-        ```
-      - The instructions JSON includes:
-        - `context`: Project background (constraints for you - do NOT include in output)
-        - `rules`: Artifact-specific rules (constraints for you - do NOT include in output)
-        - `template`: The structure to use for your output file
-        - `instruction`: Schema-specific guidance for this artifact type
-        - `resolvedOutputPath`: Resolved path or pattern to write the artifact
-        - `dependencies`: Completed artifacts to read for context
-      - Read any completed dependency files for context
-      - Create the artifact file using `template` as the structure and write it to `resolvedOutputPath`
-      - Apply `context` and `rules` as constraints - but do NOT copy them into the file
-      - Show brief progress: "Created <artifact-id>"
-
-   b. **Continue until all `applyRequires` artifacts are complete**
-      - After creating each artifact, re-run `openspec status --change "<name>" --json`
-      - Check if every artifact ID in `applyRequires` has `status: "done"` in the artifacts array
-      - Stop when all `applyRequires` artifacts are done
-
-   c. **If an artifact requires user input** (unclear context):
-      - Use **AskUserQuestion tool** to clarify
-      - Then continue with creation
-
-5. **Show final status**
-   ```bash
-   openspec status --change "<name>"
-   ```
-
-**Output**
-
-After completing all artifacts, summarize:
-- Change name and location
-- List of artifacts created with brief descriptions
-- What's ready: "All artifacts created! Ready for implementation."
-- Prompt: "Run `/opsx:apply` or ask me to implement to start working on the tasks."
-
-**Artifact Creation Guidelines**
-
-- Follow the `instruction` field from `openspec instructions` for each artifact type
-- The schema defines what each artifact should contain - follow it
-- Read dependency artifacts for context before creating new ones
-- Use `template` as the structure for your output file - fill in its sections
-- **IMPORTANT**: `context` and `rules` are constraints for YOU, not content for the file
-  - Do NOT copy `<context>`, `<rules>`, `<project_context>` blocks into the artifact
-  - These guide what you write, but should never appear in the output
-
-**Guardrails**
-- Create ALL artifacts needed for implementation (as defined by schema's `apply.requires`)
-- Always read dependency artifacts before creating a new one
-- If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
-- If a change with that name already exists, ask if user wants to continue it or create a new one
-- Verify each artifact file exists after writing before proceeding to next
diff --git a/packages/.github/skills/openspec-sync-specs/SKILL.md b/packages/.github/skills/openspec-sync-specs/SKILL.md
deleted file mode 100644
index b6aad67a..00000000
--- a/packages/.github/skills/openspec-sync-specs/SKILL.md
+++ /dev/null
@@ -1,148 +0,0 @@
----
-name: openspec-sync-specs
-description: Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Sync delta specs from a change to main specs.
-
-This is an **agent-driven** operation - you will read delta specs and directly edit main specs to apply the changes. This allows intelligent merging (e.g., adding a scenario without copying the entire requirement).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have delta specs (under `specs/` directory).
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Resolve change context**
-
-   Run:
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-
-3. **Find delta specs**
-
-   Use `artifactPaths.specs.existingOutputPaths` from the status JSON as the list of delta spec files.
-
-   Each delta spec file contains sections like:
-   - `## ADDED Requirements` - New requirements to add
-   - `## MODIFIED Requirements` - Changes to existing requirements
-   - `## REMOVED Requirements` - Requirements to remove
-   - `## RENAMED Requirements` - Requirements to rename (FROM:/TO: format)
-
-   If no delta specs found, inform user and stop.
-
-4. **For each delta spec, apply changes to main specs**
-
-   For each repo-local capability delta spec path returned by the CLI:
-
-   a. **Read the delta spec** to understand the intended changes
-
-   b. **Read the main spec** at `openspec/specs/<capability>/spec.md` (may not exist yet)
-
-   c. **Apply changes intelligently**:
-
-      **ADDED Requirements:**
-      - If requirement doesn't exist in main spec → add it
-      - If requirement already exists → update it to match (treat as implicit MODIFIED)
-
-      **MODIFIED Requirements:**
-      - Find the requirement in main spec
-      - Apply the changes - this can be:
-        - Adding new scenarios (don't need to copy existing ones)
-        - Modifying existing scenarios
-        - Changing the requirement description
-      - Preserve scenarios/content not mentioned in the delta
-
-      **REMOVED Requirements:**
-      - Remove the entire requirement block from main spec
-
-      **RENAMED Requirements:**
-      - Find the FROM requirement, rename to TO
-
-   d. **Create new main spec** if capability doesn't exist yet:
-      - Create `openspec/specs/<capability>/spec.md`
-      - Add Purpose section (can be brief, mark as TBD)
-      - Add Requirements section with the ADDED requirements
-
-5. **Show summary**
-
-   After applying all changes, summarize:
-   - Which capabilities were updated
-   - What changes were made (requirements added/modified/removed/renamed)
-
-**Delta Spec Format Reference**
-
-```markdown
-## ADDED Requirements
-
-### Requirement: New Feature
-The system SHALL do something new.
-
-#### Scenario: Basic case
-- **WHEN** user does X
-- **THEN** system does Y
-
-## MODIFIED Requirements
-
-### Requirement: Existing Feature
-#### Scenario: New scenario to add
-- **WHEN** user does A
-- **THEN** system does B
-
-## REMOVED Requirements
-
-### Requirement: Deprecated Feature
-
-## RENAMED Requirements
-
-- FROM: `### Requirement: Old Name`
-- TO: `### Requirement: New Name`
-```
-
-**Key Principle: Intelligent Merging**
-
-Unlike programmatic merging, you can apply **partial updates**:
-- To add a scenario, just include that scenario under MODIFIED - don't copy existing scenarios
-- The delta represents *intent*, not a wholesale replacement
-- Use your judgment to merge changes sensibly
-
-**Output On Success**
-
-```
-## Specs Synced: <change-name>
-
-Updated main specs:
-
-**<capability-1>**:
-- Added requirement: "New Feature"
-- Modified requirement: "Existing Feature" (added 1 scenario)
-
-**<capability-2>**:
-- Created new spec file
-- Added requirement: "Another Feature"
-
-Main specs are now updated. The change remains active - archive when implementation is complete.
-```
-
-**Guardrails**
-- Read both delta and main specs before making changes
-- Preserve existing content not mentioned in delta
-- If something is unclear, ask for clarification
-- Show what you're changing as you go
-- The operation should be idempotent - running twice should give same result
diff --git a/packages/.github/skills/openspec-update-change/SKILL.md b/packages/.github/skills/openspec-update-change/SKILL.md
deleted file mode 100644
index 3e1c931d..00000000
--- a/packages/.github/skills/openspec-update-change/SKILL.md
+++ /dev/null
@@ -1,86 +0,0 @@
----
-name: openspec-update-change
-description: Update an OpenSpec change by revising its existing planning artifacts and keeping them coherent with one another. Use when the user wants to revise a change's plan, fold new decisions into it, or reconcile its artifacts after an edit. Never edits code.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Revise a change's existing planning artifacts and keep them coherent. Never edit code.
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes sorted by most recently modified. Then use the **AskUserQuestion tool** to let the user select which change to update.
-
-   Present the top 3-4 most recently modified changes as options, showing:
-   - Change name
-   - Schema (from `schema` field if present, otherwise "spec-driven")
-   - Status (e.g., "0/5 tasks", "complete", "no tasks")
-   - How recently it was modified (from `lastModified` field)
-
-   Mark the most recently modified change as "(Recommended)" since it's likely what the user wants to update.
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Get the change's artifacts**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand current state. The response includes:
-   - `schemaName`: The workflow schema being used (e.g., "spec-driven")
-   - `artifacts`: Array of artifacts with their status ("done", "ready", "blocked")
-   - `isComplete`: Boolean indicating if all artifacts are complete
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context. Use these instead of assuming repo-local paths.
-
-   The artifact ids and paths come from the active schema - do NOT assume them, and do NOT branch on hardcoded artifact names. Custom schemas must work unchanged.
-
-   The files to edit are `artifactPaths.<id>.existingOutputPaths` - the concrete files that exist on disk, already glob-expanded for glob artifacts (e.g. `specs/**/*.md`). Do NOT write to `resolvedOutputPath`: for a glob artifact it is still the glob pattern, not a real file.
-
-3. **Understand the request**
-   - If the user asked for a specific revision ("the design now uses X"), that is the starting edit.
-   - If they only said "update" / "make this coherent", treat it as a coherence review: read the existing artifacts and check them against each other for contradictions, gaps, and duplication.
-
-4. **Read and reconcile**
-   - Read the artifact(s) the request touches and the change's other existing artifacts.
-   - Apply the requested edit. Then check every other existing artifact against it - in ANY direction: an edit to a later artifact may require revising an earlier one, not only the other way around. Build order is a useful reading order, not a constraint on which artifacts may be revised.
-   - Note everything that is now inconsistent, missing, or contradictory.
-   - Revise only files that already exist (`existingOutputPaths`). Do NOT create artifacts that don't exist yet, and do NOT invent new files under a glob artifact - note them and point the user to `/opsx:continue` to create them.
-   - If the change is already coherent, say so and make no edits.
-
-5. **Confirm and apply, one artifact at a time**
-   - Show each proposed revision and why. Write only after the user confirms.
-   - If the user rejects a revision, do not write it - leave that artifact unchanged.
-   - When a substantial rewrite is needed, get that artifact's rules and template first:
-     ```bash
-     openspec instructions <artifact-id> --change "<name>" --json
-     ```
-
-6. **Point to the next step (guidance only - NEVER act on it)**
-   - Artifacts still missing -> suggest `/opsx:continue` to create them.
-   - Change already implemented (tasks checked off / already applied) -> the code may no longer match the revised plan; suggest `/opsx:apply` to carry the delta into code.
-   - Everything done and implemented -> suggest `/opsx:archive`.
-
-**Output**
-
-After each invocation, show:
-- Which artifacts were revised (and which proposed revisions were rejected)
-- Anything deferred to `/opsx:continue` (not-yet-created artifacts or files)
-- Where the change stands and the recommended next command
-
-**Guardrails**
-- Planning artifacts only - NEVER edit implementation code. If the revised plan implies code changes, stop and point to `/opsx:apply`.
-- Use the artifact ids and paths reported by `openspec status`; never branch on hardcoded artifact names.
-- Edit only the concrete files in `existingOutputPaths`; never write to a glob `resolvedOutputPath`.
-- Do not advance the build frontier: no new artifacts, no new files under glob artifacts - that is `/opsx:continue`'s job.
-- Confirm every edit with the user before writing.
-- If the request changes the change's *intent* rather than refining it, recommend starting fresh with `/opsx:new` (the "Update vs. Start Fresh" heuristic).
diff --git a/packages/.github/skills/openspec-verify-change/SKILL.md b/packages/.github/skills/openspec-verify-change/SKILL.md
deleted file mode 100644
index b6557c6f..00000000
--- a/packages/.github/skills/openspec-verify-change/SKILL.md
+++ /dev/null
@@ -1,172 +0,0 @@
----
-name: openspec-verify-change
-description: Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
-allowed-tools: Bash(openspec:*)
-license: MIT
-compatibility: Requires openspec CLI.
-metadata:
-  author: openspec
-  version: "1.0"
-  generatedBy: "1.6.0"
----
-
-Verify that an implementation matches the change artifacts (specs, tasks, design).
-
-**Store selection:** If the user names a store (a store is a standalone OpenSpec repo registered on this machine) or the work lives in one, run `openspec store list --json` to discover registered store ids, then pass `--store <id>` on the commands that read or write specs and changes (`new change`, `status`, `instructions`, `list`, `show`, `validate`, `archive`, `doctor`, `context`). Other commands do not take the flag. Hints printed by commands already carry the flag; keep it on follow-ups. Without a store, commands act on the nearest local `openspec/` root.
-
-**Input**: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
-
-**Steps**
-
-1. **If no change name provided, prompt for selection**
-
-   Run `openspec list --json` to get available changes. Use the **AskUserQuestion tool** to let the user select.
-
-   Show changes that have implementation tasks (tasks artifact exists).
-   Include the schema used for each change if available.
-   Mark changes with incomplete tasks as "(In Progress)".
-
-   **IMPORTANT**: Do NOT guess or auto-select a change. Always let the user choose.
-
-2. **Check status to understand the schema**
-   ```bash
-   openspec status --change "<name>" --json
-   ```
-   Parse the JSON to understand:
-   - `schemaName`: The workflow being used (e.g., "spec-driven")
-   - `planningHome`, `changeRoot`, `artifactPaths`, and `actionContext`: path and scope context
-   - Which artifacts exist for this change
-
-3. **Get planning context and load artifacts**
-
-   ```bash
-   openspec instructions apply --change "<name>" --json
-   ```
-
-   This returns the change directory and `contextFiles` (artifact ID -> array of concrete file paths). Read all available artifacts from `contextFiles`.
-
-4. **Initialize verification report structure**
-
-   Create a report structure with three dimensions:
-   - **Completeness**: Track tasks and spec coverage
-   - **Correctness**: Track requirement implementation and scenario coverage
-   - **Coherence**: Track design adherence and pattern consistency
-
-   Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
-5. **Verify Completeness**
-
-   **Task Completion**:
-   - If `contextFiles.tasks` exists, read every file path in it
-   - Parse checkboxes: `- [ ]` (incomplete) vs `- [x]` (complete)
-   - Count complete vs total tasks
-   - If incomplete tasks exist:
-     - Add CRITICAL issue for each incomplete task
-     - Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
-
-   **Spec Coverage**:
-   - If delta specs exist in `contextFiles.specs`:
-     - Extract all requirements (marked with "### Requirement:")
-     - For each requirement:
-       - Search codebase for keywords related to the requirement
-       - Assess if implementation likely exists
-     - If requirements appear unimplemented:
-       - Add CRITICAL issue: "Requirement not found: <requirement name>"
-       - Recommendation: "Implement requirement X: <description>"
-
-6. **Verify Correctness**
-
-   **Requirement Implementation Mapping**:
-   - For each requirement from delta specs:
-     - Search codebase for implementation evidence
-     - If found, note file paths and line ranges
-     - Assess if implementation matches requirement intent
-     - If divergence detected:
-       - Add WARNING: "Implementation may diverge from spec: <details>"
-       - Recommendation: "Review <file>:<lines> against requirement X"
-
-   **Scenario Coverage**:
-   - For each scenario in delta specs (marked with "#### Scenario:"):
-     - Check if conditions are handled in code
-     - Check if tests exist covering the scenario
-     - If scenario appears uncovered:
-       - Add WARNING: "Scenario not covered: <scenario name>"
-       - Recommendation: "Add test or implementation for scenario: <description>"
-
-7. **Verify Coherence**
-
-   **Design Adherence**:
-   - If `contextFiles.design` exists:
-     - Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
-     - Verify implementation follows those decisions
-     - If contradiction detected:
-       - Add WARNING: "Design decision not followed: <decision>"
-       - Recommendation: "Update implementation or revise design.md to match reality"
-   - If no design.md: Skip design adherence check, note "No design.md to verify against"
-
-   **Code Pattern Consistency**:
-   - Review new code for consistency with project patterns
-   - Check file naming, directory structure, coding style
-   - If significant deviations found:
-     - Add SUGGESTION: "Code pattern deviation: <details>"
-     - Recommendation: "Consider following project pattern: <example>"
-
-8. **Generate Verification Report**
-
-   **Summary Scorecard**:
-   ```
-   ## Verification Report: <change-name>
-
-   ### Summary
-   | Dimension    | Status           |
-   |--------------|------------------|
-   | Completeness | X/Y tasks, N reqs|
-   | Correctness  | M/N reqs covered |
-   | Coherence    | Followed/Issues  |
-   ```
-
-   **Issues by Priority**:
-
-   1. **CRITICAL** (Must fix before archive):
-      - Incomplete tasks
-      - Missing requirement implementations
-      - Each with specific, actionable recommendation
-
-   2. **WARNING** (Should fix):
-      - Spec/design divergences
-      - Missing scenario coverage
-      - Each with specific recommendation
-
-   3. **SUGGESTION** (Nice to fix):
-      - Pattern inconsistencies
-      - Minor improvements
-      - Each with specific recommendation
-
-   **Final Assessment**:
-   - If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
-   - If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
-   - If all clear: "All checks passed. Ready for archive."
-
-**Verification Heuristics**
-
-- **Completeness**: Focus on objective checklist items (checkboxes, requirements list)
-- **Correctness**: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
-- **Coherence**: Look for glaring inconsistencies, don't nitpick style
-- **False Positives**: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
-- **Actionability**: Every issue must have a specific recommendation with file/line references where applicable
-
-**Graceful Degradation**
-
-- If only tasks.md exists: verify task completion only, skip spec/design checks
-- If tasks + specs exist: verify completeness and correctness, skip design
-- If full artifacts: verify all three dimensions
-- Always note which checks were skipped and why
-
-**Output Format**
-
-Use clear markdown with:
-- Table for summary scorecard
-- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
-- Code references in format: `file.ts:123`
-- Specific, actionable recommendations
-- No vague suggestions like "consider reviewing"
diff --git a/packages/openspec/config.yaml b/packages/openspec/config.yaml
deleted file mode 100644
index 392946c6..00000000
--- a/packages/openspec/config.yaml
+++ /dev/null
@@ -1,20 +0,0 @@
-schema: spec-driven
-
-# Project context (optional)
-# This is shown to AI when creating artifacts.
-# Add your tech stack, conventions, style guides, domain knowledge, etc.
-# Example:
-#   context: |
-#     Tech stack: TypeScript, React, Node.js
-#     We use conventional commits
-#     Domain: e-commerce platform
-
-# Per-artifact rules (optional)
-# Add custom rules for specific artifacts.
-# Example:
-#   rules:
-#     proposal:
-#       - Keep proposals under 500 words
-#       - Always include a "Non-goals" section
-#     tasks:
-#       - Break tasks into chunks of max 2 hours

From 658d0e9670e9a79e1691f66593c096844a8e8c6b Mon Sep 17 00:00:00 2001
From: gohabereg <gohabereg@gmail.com>
Date: Sat, 18 Jul 2026 23:16:11 +0100
Subject: [PATCH 5/5] Document required WSS_PORT env var for docker compose up

Reviewer noted docker compose up fails without WSS_PORT set explicitly.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
---
 CONTRIBUTING.md | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 4f37d580..266bed9c 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -40,7 +40,10 @@ To try changes end-to-end, use the Vite playground:
 cd packages/playground && yarn dev
 ```
 
-The OT collaboration server can be run via Docker: `docker compose up`.
+The OT collaboration server can be run via Docker. It requires a `WSS_PORT` env var (defaults to `8080` if you set it, but `docker compose` needs it explicit) — create a `.env` file at the repo root with `WSS_PORT=8080`, then:
+```bash
+docker compose up
+```
 
 ## Project structure