agentleFS
healthier-vitaminshealthier-vitamins/agentic-grimoire/skills/review-all/SKILL.md

review-all

Skill2 starsChanged 12 days agoMITView on GitHub

Run code-review, ponytail-review and (when the diff crosses a trust boundary) security-review on one diff, in one command.

7 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

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

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run the review skills below on one shared diff and report each one's findings.

    line 8
  2. Distinctive:

    Done when: every reviewer has returned a report or a recorded skip.

    line 26

“Common” is measured: kept by at least 6 in 10 of the 6 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

Run the review skills below on one shared diff and report each one's findings. This is review only. Leave the code untouched.

1. Pin the diff

  • Fixed point: the argument, else the default branch (main).
  • Diff command: git diff $(git merge-base <fixed point> HEAD). It compares the merge-base to the working tree, so uncommitted work counts. Pass this exact command to every reviewer in place of its own default.
  • Spec: the repo's plan file (PLAN.md, or the spec the commits name), if one exists.

Done when: the fixed point resolves, the diff is non-empty, and you hold the diff command, git diff --stat and the spec path.

2. Run the reviewers

Invoke each with the Skill tool, one after another, giving each the diff command:

  1. code-review: pass the fixed point and the spec path.
  2. ponytail:ponytail-review.
  3. security-review: only when the diff touches a trust boundary: auth or sign-in, secrets or tokens, permissions or tool gates, or input from outside (web pages, chat messages, HTTP requests, tool results, files the user uploads). Otherwise record "skipped: no trust boundary in the diff" and the files you checked.

Done when: every reviewer has returned a report or a recorded skip.

3. Report

One ## heading per reviewer, its findings as it gave them. Keep the reviewers separate: a finding two reviewers raised stays under both, marked "(also in X)". End with one line per reviewer naming its worst finding, then the offer: "Apply the fixes? I'll use receiving-code-review if it is installed."

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.