Skip to content

Commit 59b1f5c

Browse files
authored
Merge pull request #82 from AnExiledDev/refactor/plugin-cleanup
refactor(plugins): archive unused agents, skills, and disable 5 plugins
2 parents 047d832 + 8f97ef7 commit 59b1f5c

127 files changed

Lines changed: 178 additions & 27453 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

container/.devcontainer/AGENTS.md

Lines changed: 14 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -46,23 +46,28 @@ Config files deploy via `defaults/codeforge/file-manifest.json` on every contain
4646

4747
Declared in `settings.json` under `enabledPlugins`, auto-activated on start:
4848

49-
- **agent-system** — 19 custom agents + built-in agent redirection
50-
- **skill-engine** — 23 general coding skills + auto-suggestion
51-
- **spec-workflow**3 spec lifecycle skills (`/spec`, `/build`, `/specs`) + spec-reminder hook
52-
- **session-context**Git state injection, TODO harvesting, commit reminders
49+
### Active
50+
51+
- **agent-system**4 custom agents (architect, claude-guide, explorer, generalist) + built-in agent redirection
52+
- **skill-engine**2 coding knowledge packs (`/team`, `/agent-browser`) + auto-suggestion
5353
- **auto-code-quality** — Auto-format + auto-lint + advisory test runner
54+
- **session-context** — Git state injection, TODO harvesting, commit reminders
5455
- **workspace-scope-guard** — Blocks writes outside working directory
5556
- **dangerous-command-blocker** — Blocks destructive bash commands
5657
- **protected-files-guard** — Blocks edits to secrets/lock files
57-
- **codeforge-lsp** — LSP for Python + TypeScript/JavaScript
58-
- **ticket-workflow** — EARS ticket workflow + auto-linking
59-
- **git-workflow** — Standalone ship (commit/push/PR) + PR review
60-
- **notify-hook** — Desktop notifications on completion
6158
- **frontend-design** (Anthropic official) — UI/frontend design skill
6259
- **code-review** (Anthropic official) — Code review skill
6360
- **feature-dev** (Anthropic official) — Feature development skill
6461
- **pr-review-toolkit** (Anthropic official) — PR review commands + agents
65-
- **prompt-snippets** — Quick behavioral mode switches via /ps command
62+
63+
### Disabled
64+
65+
- **spec-workflow** — Spec lifecycle (archived, pending rewrite)
66+
- **git-workflow** — Ship + PR review (archived, pending rewrite)
67+
- **ticket-workflow** — EARS ticket workflow (archived, pending rewrite)
68+
- **notify-hook** — Desktop notifications (archived, pending rewrite)
69+
- **prompt-snippets** — Behavioral mode switches (archived, pending rewrite)
70+
- **codeforge-lsp** — LSP servers (disabled)
6671

6772
## Rules System
6873

container/.devcontainer/CHANGELOG.md

Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -20,6 +20,8 @@
2020

2121
### Configuration
2222

23+
- **Dangerous-mode permission prompt skipped by default**`skipDangerousModePermissionPrompt: true` is now set in `settings.base.json` and propagates to all five generated profiles. Suppresses the one-time bypass-permissions confirmation on new devcontainers.
24+
- **Effort level bumped to `max` on opus-4-7** — both opus-4-7 overlays (200k and 1M-400k) now set `effortLevel: "max"` and `CLAUDE_CODE_EFFORT_LEVEL: "max"`. Opus-4-5 and opus-4-6 profiles no longer carry any effort-level setting; they use `MAX_THINKING_TOKENS: 31999` with adaptive thinking disabled (token budgets, not effort levels). `CLAUDE_CODE_EFFORT_LEVEL` was removed from base settings so it no longer leaks into non-4.7 profiles.
2325
- **Claude settings profiles** — replaced the single hand-edited default with `settings.base.json` plus model overlays that generate five deployed settings files: opus-4-7 200k default, opus-4-7 1M bounded to 400k, opus-4-6 200k, opus-4-6 1M bounded to 400k, and opus-4-5 200k.
2426
- **Profile aliases**`cc`, `claude`, `cc7`, `ccw`, `ccw7`, `cc-orc`, and `cc-orc7` now use the opus-4-7 200k settings profile. Added `cc5`, `cc6`, `cc61`, `cc71` plus matching `ccw*` and `cc-orc*` variants.
2527
- **Settings-based context bounds** — Claude launchers now pass `--settings` profile files instead of inline context env vars or `--model`, so model, context, and thinking controls stay in settings JSON.
@@ -48,6 +50,19 @@
4850
- **Cross-references added** — before-you-install, accessing-services, troubleshooting, and devcontainer-cli pages now link to the Windows networking guide
4951
- **Agent-browser CDP troubleshooting** — new troubleshooting section for host Chrome connection issues (localhost vs host.docker.internal, Chrome version requirements, port exposure)
5052

53+
### Agent System
54+
55+
- **All agents and skills set to `effort: max`** — active agents (claude-guide, explorer, generalist), all 14 archived agents in `agent-system/agents/_archived/`, both active skill-engine skills (`team`, `agent-browser`), all 22 archived skills across `skill-engine` and `agent-system`, and all 10 skills in currently disabled plugins (`git-workflow`, `prompt-snippets`, `spec-workflow`, `ticket-workflow`) now declare `effort: max`. This ensures any plugin re-enable in the future yields max-effort runs without further edits.
56+
57+
### Plugin Cleanup
58+
59+
- **Archived 15 agents** — agent-system reduced from 19 to 4 active agents (architect, claude-guide, explorer, generalist). Archived agents preserved in `agents/_archived/` for future rewrite.
60+
- **Archived 21 skills** — skill-engine reduced from 23 to 2 active skills (`/team`, `/agent-browser`). Archived skills preserved in `skills/_archived/`.
61+
- **Archived `/debug` skill** — moved from agent-system active skills to `skills/_archived/`.
62+
- **Removed 2 agent redirects**`Bash→bash-exec` and `statusline-setup→statusline-config` removed from redirect map. Built-in types now fall through to Claude Code defaults.
63+
- **Disabled prompt-snippets plugin**`/ps` command no longer available.
64+
- **Stripped skill-suggester** — auto-suggestion now only covers `team` and `agent-browser` (was 25+ skills).
65+
5166
### CI
5267

5368
- **Canary pre-release publishing** — every push to `staging` that touches `container/` now auto-publishes a canary build to npm. Install with `npm i @coredirective/cf-container@canary` to try unreleased changes. Versions use the format `{version}-staging.{sha7}`.

container/.devcontainer/README.md

Lines changed: 3 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -358,29 +358,18 @@ All methods persist across container rebuilds via the bind-mounted `/workspaces/
358358

359359
Agents and skills are distributed across focused plugins (replacing the former `code-directive` monolith).
360360

361-
### Custom Agents (17) — `agent-system` plugin
361+
### Custom Agents (4) — `agent-system` plugin
362362

363363
Agent definitions in `plugins/devs-marketplace/plugins/agent-system/agents/` provide enhanced behavior when spawned via the `Task` tool. The `redirect-builtin-agents.py` hook transparently swaps built-in agent types to these custom agents.
364364

365365
| Agent | Purpose |
366366
|-------|---------|
367367
| `architect` | System design and implementation planning |
368-
| `bash-exec` | Command execution specialist |
369368
| `claude-guide` | Claude Code feature guidance |
370-
| `debug-logs` | Log analysis and error diagnosis |
371-
| `dependency-analyst` | Dependency analysis and upgrades |
372-
| `documenter` | Documentation, specs, and spec lifecycle |
373369
| `explorer` | Fast codebase search and navigation |
374370
| `generalist` | General-purpose multi-step tasks |
375-
| `git-archaeologist` | Git history forensics |
376-
| `migrator` | Code migration and upgrades |
377-
| `perf-profiler` | Performance profiling |
378-
| `refactorer` | Code refactoring with regression checks |
379-
| `researcher` | Research and information gathering |
380-
| `security-auditor` | Security vulnerability analysis |
381-
| `spec-writer` | Specification and requirements authoring |
382-
| `statusline-config` | ccstatusline configuration |
383-
| `test-writer` | Test authoring with pass verification |
371+
372+
15 previously active agents have been archived to `agents/_archived/`.
384373

385374
### General Skills (23) — `skill-engine` plugin
386375

container/.devcontainer/defaults/codeforge/config/orchestrator-system-prompt.md

Lines changed: 2 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -68,7 +68,7 @@ You are a delegation-first orchestrator. You decompose tasks, delegate to agents
6868

6969
Hard rules:
7070
- NEVER use `Edit` or `Write` tools — delegate to the implementer or documenter agent
71-
- NEVER use `Bash` for commands with side effects — delegate to the implementer or bash-exec agent
71+
- NEVER use `Bash` for commands with side effects — delegate to the implementer agent
7272
- `Read`, `Glob`, `Grep` are permitted for quick context gathering before delegation
7373
- NEVER write code, generate patches, or produce implementation artifacts directly
7474
- NEVER run tests directly — delegate to the tester agent
@@ -140,9 +140,7 @@ Specialist agents (use when a workhorse doesn't fit):
140140
|-------|--------|--------|-------|---------|
141141
| architect | Architecture planning | Read-only | Opus | Complex system design, trade-off analysis, implementation planning |
142142
| security-auditor | Security | Read-only | Sonnet | OWASP audits, secrets scanning, vulnerability detection |
143-
| bash-exec | Command execution | Bash only | Sonnet | Simple terminal commands when no other agent is appropriate |
144143
| claude-guide | Claude Code help | Read-only | Haiku | Claude Code features, configuration, SDK questions |
145-
| statusline-config | Status line | Read-write | Sonnet | Claude Code status line widget configuration |
146144

147145
Selection criteria:
148146
- Is the task research/investigation? → investigator
@@ -151,8 +149,7 @@ Selection criteria:
151149
- Does the task involve documentation or specs? → documenter
152150
- Is it a targeted security review? → security-auditor
153151
- Is it a complex architecture decision? → architect
154-
- Is it a simple command to run? → bash-exec
155-
- Does the task require a specialist not listed above? → consult the agent-system README for the full 17-agent specialist catalog
152+
- Does the task require a specialist not listed above? → consult the agent-system README for the full agent catalog
156153
</agent_catalog>
157154

158155
<question_surfacing>

container/.devcontainer/defaults/codeforge/config/settings-opus-45-200k.json

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,14 +2,14 @@
22
"cleanupPeriodDays": 60,
33
"autoCompact": true,
44
"alwaysThinkingEnabled": true,
5+
"skipDangerousModePermissionPrompt": true,
56
"env": {
67
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-4-6",
78
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
89
"BASH_DEFAULT_TIMEOUT_MS": "120000",
910
"BASH_MAX_TIMEOUT_MS": "300000",
1011
"CLAUDE_CODE_MAX_OUTPUT_TOKENS": "64000",
1112
"MAX_MCP_OUTPUT_TOKENS": "10000",
12-
"CLAUDE_CODE_EFFORT_LEVEL": "high",
1313
"MCP_TIMEOUT": "120000",
1414
"MCP_TOOL_TIMEOUT": "30000",
1515
"CLAUDE_AUTOCOMPACT_PCT_OVERRIDE": "80",
@@ -121,7 +121,7 @@
121121
"session-context@devs-marketplace": false,
122122
"auto-code-quality@devs-marketplace": true,
123123
"workspace-scope-guard@devs-marketplace": true,
124-
"prompt-snippets@devs-marketplace": true,
124+
"prompt-snippets@devs-marketplace": false,
125125
"git-workflow@devs-marketplace": false
126126
},
127127
"autoUpdatesChannel": "latest",

container/.devcontainer/defaults/codeforge/config/settings-opus-46-1m-400k.json

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,14 +2,14 @@
22
"cleanupPeriodDays": 60,
33
"autoCompact": true,
44
"alwaysThinkingEnabled": true,
5+
"skipDangerousModePermissionPrompt": true,
56
"env": {
67
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-4-6",
78
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
89
"BASH_DEFAULT_TIMEOUT_MS": "120000",
910
"BASH_MAX_TIMEOUT_MS": "300000",
1011
"CLAUDE_CODE_MAX_OUTPUT_TOKENS": "64000",
1112
"MAX_MCP_OUTPUT_TOKENS": "10000",
12-
"CLAUDE_CODE_EFFORT_LEVEL": "high",
1313
"MCP_TIMEOUT": "120000",
1414
"MCP_TOOL_TIMEOUT": "30000",
1515
"CLAUDE_AUTOCOMPACT_PCT_OVERRIDE": "80",
@@ -121,7 +121,7 @@
121121
"session-context@devs-marketplace": false,
122122
"auto-code-quality@devs-marketplace": true,
123123
"workspace-scope-guard@devs-marketplace": true,
124-
"prompt-snippets@devs-marketplace": true,
124+
"prompt-snippets@devs-marketplace": false,
125125
"git-workflow@devs-marketplace": false
126126
},
127127
"autoUpdatesChannel": "latest",

container/.devcontainer/defaults/codeforge/config/settings-opus-46-200k.json

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,14 +2,14 @@
22
"cleanupPeriodDays": 60,
33
"autoCompact": true,
44
"alwaysThinkingEnabled": true,
5+
"skipDangerousModePermissionPrompt": true,
56
"env": {
67
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-4-6",
78
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
89
"BASH_DEFAULT_TIMEOUT_MS": "120000",
910
"BASH_MAX_TIMEOUT_MS": "300000",
1011
"CLAUDE_CODE_MAX_OUTPUT_TOKENS": "64000",
1112
"MAX_MCP_OUTPUT_TOKENS": "10000",
12-
"CLAUDE_CODE_EFFORT_LEVEL": "high",
1313
"MCP_TIMEOUT": "120000",
1414
"MCP_TOOL_TIMEOUT": "30000",
1515
"CLAUDE_AUTOCOMPACT_PCT_OVERRIDE": "80",
@@ -121,7 +121,7 @@
121121
"session-context@devs-marketplace": false,
122122
"auto-code-quality@devs-marketplace": true,
123123
"workspace-scope-guard@devs-marketplace": true,
124-
"prompt-snippets@devs-marketplace": true,
124+
"prompt-snippets@devs-marketplace": false,
125125
"git-workflow@devs-marketplace": false
126126
},
127127
"autoUpdatesChannel": "latest",

container/.devcontainer/defaults/codeforge/config/settings-opus-47-1m-400k.json

Lines changed: 5 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -2,14 +2,14 @@
22
"cleanupPeriodDays": 60,
33
"autoCompact": true,
44
"alwaysThinkingEnabled": true,
5+
"skipDangerousModePermissionPrompt": true,
56
"env": {
67
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-4-6",
78
"ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
89
"BASH_DEFAULT_TIMEOUT_MS": "120000",
910
"BASH_MAX_TIMEOUT_MS": "300000",
1011
"CLAUDE_CODE_MAX_OUTPUT_TOKENS": "64000",
1112
"MAX_MCP_OUTPUT_TOKENS": "10000",
12-
"CLAUDE_CODE_EFFORT_LEVEL": "high",
1313
"MCP_TIMEOUT": "120000",
1414
"MCP_TOOL_TIMEOUT": "30000",
1515
"CLAUDE_AUTOCOMPACT_PCT_OVERRIDE": "80",
@@ -36,7 +36,8 @@
3636
"ANTHROPIC_MODEL": "claude-opus-4-7[1m]",
3737
"ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4-7[1m]",
3838
"CLAUDE_CODE_MAX_CONTEXT_TOKENS": "400000",
39-
"CLAUDE_CODE_AUTO_COMPACT_WINDOW": "400000"
39+
"CLAUDE_CODE_AUTO_COMPACT_WINDOW": "400000",
40+
"CLAUDE_CODE_EFFORT_LEVEL": "max"
4041
},
4142
"teammateMode": "auto",
4243
"showClearContextOnPlanAccept": true,
@@ -119,11 +120,11 @@
119120
"session-context@devs-marketplace": false,
120121
"auto-code-quality@devs-marketplace": true,
121122
"workspace-scope-guard@devs-marketplace": true,
122-
"prompt-snippets@devs-marketplace": true,
123+
"prompt-snippets@devs-marketplace": false,
123124
"git-workflow@devs-marketplace": false
124125
},
125126
"autoUpdatesChannel": "latest",
126127
"model": "claude-opus-4-7[1m]",
127128
"autoCompactWindow": 400000,
128-
"effortLevel": "high"
129+
"effortLevel": "max"
129130
}
Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,12 +1,12 @@
11
{
22
"model": "claude-opus-4-7[1m]",
33
"autoCompactWindow": 400000,
4-
"effortLevel": "high",
4+
"effortLevel": "max",
55
"env": {
66
"ANTHROPIC_MODEL": "claude-opus-4-7[1m]",
77
"ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4-7[1m]",
88
"CLAUDE_CODE_MAX_CONTEXT_TOKENS": "400000",
99
"CLAUDE_CODE_AUTO_COMPACT_WINDOW": "400000",
10-
"CLAUDE_CODE_EFFORT_LEVEL": "high"
10+
"CLAUDE_CODE_EFFORT_LEVEL": "max"
1111
}
1212
}
Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,12 +1,12 @@
11
{
22
"model": "claude-opus-4-7",
33
"autoCompactWindow": 200000,
4-
"effortLevel": "high",
4+
"effortLevel": "max",
55
"env": {
66
"ANTHROPIC_MODEL": "claude-opus-4-7",
77
"ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4-7",
88
"CLAUDE_CODE_MAX_CONTEXT_TOKENS": "200000",
99
"CLAUDE_CODE_AUTO_COMPACT_WINDOW": "200000",
10-
"CLAUDE_CODE_EFFORT_LEVEL": "high"
10+
"CLAUDE_CODE_EFFORT_LEVEL": "max"
1111
}
1212
}

0 commit comments

Comments
 (0)