chore(hooks): wedge-proof commands + drop project tool_guard

[zackees]

Follow-up to #712.

Summary

• Drop the project-local tool_guard.py PreToolUse hook (user runs it globally).
• Every remaining hook command is now wedge-proof:
  • Prefer \$CLAUDE_PROJECT_DIR (harness anchor that survives worktree shifts) → fall back to git rev-parse --show-toplevel.
  • [ -f <script> ] && uv run ... || exit 0 so a worktree without ci/hooks/ (orphan branch, fresh checkout-in-progress) cannot block the shell.

Root cause

During the online-data orphan-branch setup in #712, git rm -rf . inside the worktree removed ci/hooks/. The PreToolUse command then resolved git rev-parse --show-toplevel to that worktree, tried uv run ci/hooks/tool_guard.py, and python errored with "no such file" — blocking every subsequent Bash/Edit/Write call. The guard turns "script missing" into a no-op instead of a block.

Test plan

• The new command shape exits 0 cleanly from both inside-the-repo and outside-the-repo cwds (smoke-tested locally).
• No production-side behavior change — when scripts exist, they still run identically.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

@zackees, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 22 minutes and 45 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

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 credits.

🚦 How do rate limits work?

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

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, the refill rate gradually slows as usage increases. The highest same-day bursts are limited more strictly.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 2b7a1b71-fc63-49ec-b5ec-752607c40661

📥 Commits

Reviewing files that changed from the base of the PR and between 302b01c and 3943038.

📒 Files selected for processing (1)

• .claude/settings.json

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch chore/wedge-proof-hook-commands

---

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 and usage tips.}