agentleFS
Sign inStart free
loganware05loganware05/captains-compass-cursor/.cursor/skills/review-fix-loop/SKILL.md

review-fix-loop

Skill2 starsChanged 32 days agoMITView on GitHub

Iterates on PR or review feedback for a small change until tests pass and the review is clean or a human decision is required

10 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Preflight: If the PR is too large for a reliable loop, stop and propose a split.

    line 21
  2. Distinctive:

    Read the full PR diff (or git diff against the base branch).

    line 22
  3. Distinctive:

    Read the review feedback carefully.

    line 23
  4. Very specific:

    Run applicable validation (testing-validation Skill).

    line 26
  5. Distinctive:

    If more feedback arrives, repeat from step 2.

    line 28
  6. Very specific:

    Stop when the PR is clean or blocked on a Captain decision.

    line 29

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review-Fix Loop

Use this Skill when

A small PR or feature branch has actionable review feedback (human, Bugbot, Greptile-style tools, adversarial-reviewer report, or CI review comments) and the success condition is clear.

Inputs

  • Current branch / PR
  • Review feedback (paste, gh review comments, or evidence path)
  • Required end state (e.g. tests pass, review threads resolved, no unrelated rewrites)
  • Project test commands from TESTING.md

Procedure

  1. Preflight: If the PR is too large for a reliable loop, stop and propose a split. Do not start the loop on huge diffs.
  2. Read the full PR diff (or git diff against the base branch).
  3. Read the review feedback carefully.
  4. Classify each item: real/relevant, false positive, or needs human product decision.
  5. Fix only real/relevant items. Add or update tests when fixing a bug.
  6. Run applicable validation (testing-validation Skill).
  7. Summarize resolved items and remaining open questions.
  8. If more feedback arrives, repeat from step 2.
  9. Stop when the PR is clean or blocked on a Captain decision. Hand off to pull-request-preparation when ready.

Output

Updated branch with fixes, validation evidence, and a loop summary (resolved vs deferred vs needs human).

Prohibited actions

  • Auto-merging
  • Blindly accepting every review comment
  • Unrelated refactors or scope expansion
  • Continuing past autonomy budget without a Budget Stop Report
  • Treating a clean review as proof the product is valuable—only that this diff looks sound

Compose with

  • Upstream verified findings from code-reviewer
  • Supervised repair dry-run: ./scripts/northstar repair start --report … --finding … (M32/B4; never auto-merges; Captain flag required to authorize dispatch)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free