fix(bin): untrack committed hook artifacts and gitignore per-task hook files

[trillium]

Intent

fix the reused-worktree stale-hooks bug — root cause was .claude/settings.local.json accidentally COMMITTED to main in PR #20 with a dead task's hooks, so every worktree checked it out as tracked baseline; fix = untrack .claude/settings.local.json + gitignore all per-task hook artifacts (.claude/settings.local.json, .opencode/plugins/fm-busy-state.js, .fm-grok-turnend, .fm-kimi-turnend) so none can be committed again; fm-spawn hook-install was already correct; added tests/fm-hook-artifacts-not-tracked.test.sh regression guard; did not touch fm-teardown.sh per collision

What Changed

• Removed .claude/settings.local.json from tracking — it had been accidentally committed to main with a dead task's busy-hook commands, so every worktree checked it out as tracked baseline instead of getting fresh per-task hooks from fm-spawn.
• Added root .gitignore entries for all per-task, worktree-resident hook artifacts (.claude/settings.local.json, .opencode/plugins/fm-busy-state.js, .opencode/plugins/fm-turn-end.js, .fm-grok-turnend, .fm-kimi-turnend) so none of them can be committed again, with a comment explaining why a pooled worktree's next tenant would otherwise inherit a stale task's hooks.
• Added tests/fm-hook-artifacts-not-tracked.test.sh as a regression guard, and corrected a stale claim in .agents/skills/harness-adapters/SKILL.md that git info/exclude alone gitignores these files — it's now documented as a per-worktree-local backstop, with the tracked root .gitignore as the actual mechanism.

Risk Assessment

✅ Low: The change only deletes a previously-accidentally-committed per-task artifact file and adds matching .gitignore entries plus a regression test; it's non-functional repo hygiene, verified consistent with fm-spawn.sh's actual exclude_path() call sites (settings.local.json, fm-busy-state.js, .fm-grok-turnend, .fm-kimi-turnend), and the shared .claude/settings.json remains tracked.

Testing

Ran the new regression test plus the directly related fm-spawn hook-refresh test and the adjacent gitignore-contract test, all passing on the target commit; additionally reproduced the pre-fix bug in a scratch worktree checked out at the base commit (showing .claude/settings.local.json was tracked and not gitignored there), which confirms the target commit's untrack+gitignore change is a genuine, verifiable fix rather than just passing tests.

Pipeline

Updates from git push no-mistakes

✅ **intent** - passed

✅ No issues found.

✅ **Rebase** - passed

✅ No issues found.

⚠️ **Review** - 1 info

• ℹ️ .gitignore:24 - The gitignore/test list includes /.opencode/plugins/fm-turn-end.js, but bin/fm-spawn.sh's opencode hook-install writes .opencode/plugins/fm-busy-state.js, not fm-turn-end.js — there's no exclude_path() call site for fm-turn-end.js in fm-spawn.sh despite the .gitignore comment claiming to be 'in sync with the exclude_path() call sites'. It's only referenced in fm-teardown.sh's rm -f scrub list (likely a vestigial/renamed artifact). Harmless (ignoring a path that's never created doesn't hurt), but the comment is slightly inaccurate and the test's coverage for that entry is vacuous.

✅ **Test** - passed

✅ No issues found.

• bin/fm-test-run.sh tests/fm-hook-artifacts-not-tracked.test.sh (new regression guard) — pass
• bin/fm-test-run.sh tests/fm-spawn-reused-worktree-hooks.test.sh (related fm-spawn hook-refresh logic test) — pass
• bin/fm-test-run.sh tests/fm-gitignore-config.test.sh (adjacent gitignore contract test) — pass
• Manual: git worktree add --detach <tmp> 56369d05198713c05da89f18e1fcabe0340c67c4 then git check-ignore -q .claude/settings.local.json (not ignored) and git ls-files -- .claude/settings.local.json (tracked) — confirmed the bug is present at base commit, absent at target

✅ **Document** - passed

✅ No issues found.

✅ **Lint** - passed

✅ No issues found.

✅ **Push** - passed

✅ No issues found.

[coderabbitai]

Warning

Review limit reached

@trillium, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 35 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: ca45c024-bdf7-4dbf-bbec-bff8b3d9b4a2

📥 Commits

Reviewing files that changed from the base of the PR and between 56369d0 and 80554e1.

📒 Files selected for processing (4)

• .agents/skills/harness-adapters/SKILL.md
• .claude/settings.local.json
• .gitignore
• tests/fm-hook-artifacts-not-tracked.test.sh

---

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.}