CO

Executes standardized code reviews on GitHub/GitLab PRs, resulting in a single repository-based markdown artifact.

Install

mkdir -p .claude/skills/code-review-trevormil && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13518" && unzip -o skill.zip -d .claude/skills/code-review-trevormil && rm skill.zip

Installs to .claude/skills/code-review-trevormil

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Compatibility launcher for the code-review agent. Delegate a GitHub PR review to Codex via codex exec. Codex runs the test suite first (the gate), scores six axes, and writes ONE combined artifact IN THIS REPO under .reviews/<pr>/<sha>.md plus findings.json/suggestions.json. Use when the user runs /code-review, asks to review a PR, or is preparing a PR for human merge.
371 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Intermediate

Key capabilities

  • Delegate GitHub PR reviews to Codex
  • Run test suites as a gate for reviews
  • Score code quality on six axes
  • Write combined review artifacts to the repository
  • Generate findings and suggestions in JSON format

How it works

The skill delegates the PR review to Codex, which runs the test suite, scores six axes, and writes a combined artifact in the repository.

Inputs & outputs

You give it
GitHub PR URL
You get back
A combined artifact in .reviews/<pr-number>/<short_sha>.md, findings.json, and suggestions.json

When to use code-review

  • Reviewing a new pull request
  • Performing automated pre-merge checks
  • Scoring code quality against custom criteria

About code-review

Delegates repository-wide code reviews to Codex to ensure consistency. It runs existing test suites, performs a six-axis quality analysis, and writes a comprehensive report to `.reviews/`.

Compatibility launcher for the code-review agent. Delegate a GitHub PR review to Codex via codex exec. Codex runs the test suite first (the gate), scores six axes, and writes ONE combined artifact IN THIS REPO under .reviews/<pr>/<sha>.md plus findings.json/suggestions.json. Use when the user runs /

When not to use it

  • When performing the review directly inside Claude
  • When using the built-in /review command
  • When performing standalone test runs

Limitations

  • Tests must pass for a non-blocked verdict
  • Approve requires zero medium or higher findings and a one-sentence articulation of safety
  • Any push/force-push/merge to a protected branch results in a critical finding and blocked verdict

How it compares

This skill provides a standardized, in-repository review process with specific scoring and artifact generation, unlike ad-hoc reviews.

Compared to similar skills

code-review side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
code-review (this skill)02moReviewIntermediate
python-testing-patterns772moReviewIntermediate
dependency-upgrade265moReviewIntermediate
test-cases577moNo flagsBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry