fix(templates): stop instructing a second date prefix on dated archive names

[mc856]

The agent workflow templates carry the same double-dating bug as the CLI (#1309, fixed in #1316): the archive-change and bulk-archive-change instructions tell agents to unconditionally build the archive target as YYYY-MM-DD-<name>, so a change already named 2026-07-04-voice-copilot-v1 comes out as archive/2026-07-07-2026-07-04-voice-copilot-v1 — whatever the CLI does, because the behavior is baked into instruction text.

This mirrors #1316's keep-as-is rule in the templates:

• archive-change (skill + command): the generate-target-name step now says use the change name as-is when it already starts with a YYYY-MM-DD- prefix, otherwise prepend the current date; the literal mv target becomes <target-name> so an agent copying the command verbatim cannot stack dates.
• bulk-archive-change (skill + command): same target-name rule added to the execute step, mv moved to <target-name>, and the "Archive directory target uses current date" guardrail carries the caveat.
• onboard: the archived-path example in the walkthrough now shows <target-name>, with the keep-as-is rule stated inline.
• Regenerated the committed skills/ distribution and updated the pinned template hashes; a new parity test asserts every archive-instructing template carries the caveat and never reinstates the raw stacked mv target.
• opsx-archive-skill spec: the successful-archive scenario still specified the unconditional current-date rule; now carries the same keep-as-is clause cli-archive picked up in fix(archive): keep an existing date prefix instead of stacking a new one #1316.

Notes:

• The line numbers recorded in opsx archive workflow templates stack a date prefix on already-dated change names #1317 predate the recent archive work (fix: use local dates for CLI date-only values #1361/fix(archive): stop failing on specs that were already synced before archiving #1376/fix(archive): stop failing on RENAMED deltas that were already synced #1386); re-checked against current main, all flagged sites survive unchanged, plus the command-variant duplicate of the bulk guardrail (same text, fixed identically). New since the issue: feat(skills): publish workflow skills to skills.sh #1357 committed a generated skills/ mirror, so the fix also flows through pnpm generate:skills.
• Swept in the follow-up commit, per review: the single-archive summary and failure examples and the bulk per-change rows now name the derived target, and the regression guard widened from the mv target alone to any YYYY-MM-DD- appearing as a path segment. The rule statements name the prefix but never place it in a path, so they stay untouched — as do the bulk Guardrails bullets, which already state both branches. onboard wasn't named in the review; its Phase 10 **SHOW:** block is swept too because it's a printed example and the parity test's surface list already covers onboard — trivial to revert if you'd rather keep this to single and bulk.

Fixes #1317

Written and reviewed with Claude Code using claude-fable-5

Summary by CodeRabbit

• Bug Fixes
  • Prevented archive destinations from receiving duplicate YYYY-MM-DD- prefixes when the change name already includes a date.
  • Updated both single-change and bulk-archive workflows to use the corrected archive path naming consistently.
• Documentation
  • Clarified onboarding and workflow/guide instructions for how the Archived to: path is formed when names already start with YYYY-MM-DD-.
  • Updated examples/spec text to reflect the new destination path format.
• Tests
  • Refreshed template parity expectations and added coverage to ensure the “no double date stacking” guidance appears in generated instructions.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 38ba72bb-2048-4dfc-a9f4-db3cbd80a61c

📥 Commits

Reviewing files that changed from the base of the PR and between 076827d and 53dfe10.

📒 Files selected for processing (1)

• openspec/specs/opsx-archive-skill/spec.md

---

📝 Walkthrough

Walkthrough

Archive workflow templates, generated skill instructions, onboarding guidance, the archive specification, and parity tests now avoid adding a second YYYY-MM-DD- prefix when archiving already-dated change names. A patch changeset documents the correction.

Changes

Archive date-prefix deduplication

Layer / File(s)	Summary
Update archive target generation src/core/templates/workflows/archive-change.ts, src/core/templates/workflows/bulk-archive-change.ts, openspec/specs/opsx-archive-skill/spec.md	Archive and bulk-archive templates preserve existing date prefixes, derive a target-name, and use the corresponding archive path; the archive specification documents the same rule.
Align workflow guidance skills/openspec-archive-change/SKILL.md, skills/openspec-bulk-archive-change/SKILL.md, skills/openspec-onboard/SKILL.md, src/core/templates/workflows/onboard.ts	Skill and onboarding instructions document that already-prefixed names are used as-is.
Update parity validation and release notes test/core/templates/skill-templates-parity.test.ts, .changeset/fix-template-archive-date-dedup.md	Expected template hashes are updated, a regression parity test checks for the no-stacking rule, and the patch changeset records the fix.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related issues

• #863: Modifies the same /opsx:archive skill and command templates, with overlapping archive-path behavior.

Possibly related PRs

• Fission-AI/OpenSpec#451: Added related date-prefixed archive destination logic in the OPSX archive templates.
• Fission-AI/OpenSpec#455: Introduced the archive specification and date-prefixed archive-path rule updated here.
• Fission-AI/OpenSpec#527: Added related bulk-archive workflow templates modified by this change.

Suggested reviewers: tabishb, clay-good, alfred-openspec

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: preventing double date prefixes in archive template instructions.
Linked Issues check	✅ Passed	The changes match issue #1317 by preserving existing YYYY-MM-DD prefixes and updating archive/onboard templates accordingly.
Out of Scope Changes check	✅ Passed	The extra updates to committed skills, hashes, tests, and spec text are all directly tied to the archive-template behavior change.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

🧹 Nitpick comments (1)

src/core/templates/workflows/archive-change.ts (1)

80-87: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Update mock outputs to reflect <target-name>.

The new target-name derivation logic successfully prevents stacking date prefixes. However, the Output On Success mock examples further down in these templates still hardcode the old format (e.g., YYYY-MM-DD-<name> or YYYY-MM-DD-<change-1>). While the LLM will likely substitute the actual path correctly based on the mv command, updating these placeholders ensures perfect consistency and removes any speculative chance of the LLM hallucinating a double date in its summary.

• src/core/templates/workflows/archive-change.ts#L80-L87: update the mock output placeholders (e.g., around lines 106, 224, 237) to use <target-name> instead of YYYY-MM-DD-<name>.
• src/core/templates/workflows/archive-change.ts#L198-L205: apply the same placeholder update to the OPSX command template's output examples.
• src/core/templates/workflows/bulk-archive-change.ts#L133-L138: update the mock output placeholders (e.g., around lines 211, 225) to use <change-n-target-name> (or similar) instead of YYYY-MM-DD-<change-n>.
• src/core/templates/workflows/bulk-archive-change.ts#L385-L390: apply the same placeholder update to the bulk archive OPSX command template.

(Note: Updating these templates will require regenerating the skill files and updating the corresponding Vitest hashes).

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@src/core/templates/workflows/archive-change.ts` around lines 80 - 87, Update
all archive workflow mock output placeholders to use the derived target-name
values rather than hardcoded date-prefixed names: in
src/core/templates/workflows/archive-change.ts at lines 80-87 and 198-205, use
<target-name>; in src/core/templates/workflows/bulk-archive-change.ts at lines
133-138 and 385-390, use per-change placeholders such as <change-n-target-name>.
Regenerate the corresponding skill files and update the affected Vitest hashes.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@src/core/templates/workflows/archive-change.ts`:
- Around line 80-87: Update all archive workflow mock output placeholders to use
the derived target-name values rather than hardcoded date-prefixed names: in
src/core/templates/workflows/archive-change.ts at lines 80-87 and 198-205, use
<target-name>; in src/core/templates/workflows/bulk-archive-change.ts at lines
133-138 and 385-390, use per-change placeholders such as <change-n-target-name>.
Regenerate the corresponding skill files and update the affected Vitest hashes.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 8fa9e7b3-d9f0-4951-8d7e-002de945b881

📥 Commits

Reviewing files that changed from the base of the PR and between 596d6ba and e46d9f8.

📒 Files selected for processing (8)

• .changeset/fix-template-archive-date-dedup.md
• skills/openspec-archive-change/SKILL.md
• skills/openspec-bulk-archive-change/SKILL.md
• skills/openspec-onboard/SKILL.md
• src/core/templates/workflows/archive-change.ts
• src/core/templates/workflows/bulk-archive-change.ts
• src/core/templates/workflows/onboard.ts
• test/core/templates/skill-templates-parity.test.ts

[alfred-openspec]

The target-name derivation is correct, but the single and bulk success summaries still emit archive/YYYY-MM-DD- paths, so an already-dated change can still be reported with the exact stacked date this PR is meant to eliminate. Please switch those source-template examples to the derived target-name placeholders, regenerate the committed skills and hashes, and cover the output examples in the regression test.

[mc856]

Done in 076827d. The single-archive summary and failure examples and the bulk per-change rows now print the derived target name, and the committed skills and hashes are regenerated.

The regression guard widened from the mv target to any YYYY-MM-DD- appearing as a path segment. That covers the output examples while leaving the rule statements intact — they name the prefix but never put it in a path. Each of the twelve example sites fails the test when reverted on its own.

Two judgment calls, both cheap to reverse. onboard's Phase 10 SHOW: block got the same treatment even though you named only single and bulk, since it's a printed example and the parity test's surface list already includes onboard. The bulk Guardrails bullets I left alone — rule statements that already state both branches. I also updated opsx-archive-skill/spec.md, which still specified the unconditional current-date rule the first commit removed from the template.

Rebased onto main on the way through; #1392 had landed in the same test file.

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@openspec/specs/opsx-archive-skill/spec.md`:
- Line 97: Update the earlier successful-archive scenario to use <target-name>
for the archive directory instead of unconditionally constructing
archive/YYYY-MM-DD-<name>/. Preserve the existing target-name rule in the
archive behavior requirements, including retaining an existing YYYY-MM-DD-
prefix, so both scenarios share one consistent contract.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 0484e7b3-586a-46fd-b0b2-3a17ccf65d76

📥 Commits

Reviewing files that changed from the base of the PR and between e46d9f8 and 076827d.

📒 Files selected for processing (9)

• .changeset/fix-template-archive-date-dedup.md
• openspec/specs/opsx-archive-skill/spec.md
• skills/openspec-archive-change/SKILL.md
• skills/openspec-bulk-archive-change/SKILL.md
• skills/openspec-onboard/SKILL.md
• src/core/templates/workflows/archive-change.ts
• src/core/templates/workflows/bulk-archive-change.ts
• src/core/templates/workflows/onboard.ts
• test/core/templates/skill-templates-parity.test.ts

🚧 Files skipped from review as they are similar to previous changes (6)

• skills/openspec-archive-change/SKILL.md
• .changeset/fix-template-archive-date-dedup.md
• skills/openspec-bulk-archive-change/SKILL.md
• src/core/templates/workflows/archive-change.ts
• test/core/templates/skill-templates-parity.test.ts
• src/core/templates/workflows/bulk-archive-change.ts

[alfred-openspec]

The derived-target fix now covers the single, bulk, onboarding, generated-skill, and OPSX spec surfaces, including the previously stale success/error examples. I regenerated skills cleanly, passed strict spec validation plus 21 focused template/parity tests, and confirmed exact-head CI is green.