knofler/myai/github-pr-review

Claude Code and agent skill by knofler/myai

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Multi-perspective code review on a pull request. Run security, performance, and style checks in parallel. Consolidate into a single review with severity ratings. Triggers: review pr, pr review, review pull request, check pr, audit pr

Install

npx vskill@latest install knofler/myai/github-pr-review

SKILL.md

The same SKILL.md was published earlier as knofler/myai/github-pr-review.

GitHub PR Review Playbook

When to Use

  • A pull request needs code review before merging.
  • User says "review pr", "check pr", or provides a PR number/URL.
  • Part of the "ship it" workflow before merge approval.

Prerequisites

  • gh CLI installed and authenticated.
  • PR number or URL provided, or inferred from current branch.
  • Repository checked out locally with the PR branch available.

Playbook

1. Fetch PR Details
  • Run gh pr view <number> --json title,body,files,additions,deletions,commits.
  • Run gh pr diff <number> to get the full diff.
  • Note the PR author, base branch, and number of changed files.
2. Security Scan (Parallel Lane A)
  • Check for hardcoded secrets, API keys, tokens, or passwords in the diff.
  • Verify no .env files or credential files are included.
  • Check input validation on any new API endpoints.
  • Verify auth middleware on protected routes.
  • Flag SQL/NoSQL injection vectors and XSS risks.
3. Performance Check (Parallel Lane B)
  • Identify N+1 query patterns in database calls.
  • Check for missing async/await or unhandled promises.
  • Flag large synchronous operations that could block the event loop.
  • Review any new dependencies for bundle size impact.
  • Check for missing pagination on list endpoints.
4. Style and Standards Check (Parallel Lane C)
  • Read documentation/AI_RULES.md for project conventions.
  • Verify TypeScript strict mode compliance.
  • Check naming conventions (files, variables, components).
  • Verify error handling patterns match project standards.
  • Confirm test files exist for new functionality.
5. Consolidate Findings
  • Merge results from all three parallel scans.
  • Assign severity to each finding: Critical, Warning, Nit.
  • Group findings by file for readability.
  • Determine overall verdict: Approve, Approve with Nits, Request Changes.
6. Post Review
  • Run gh pr review <number> --body "<review>" --approve or --request-changes.
  • Format review as markdown with severity badges.
  • Include a summary table of all findings at the top.
7. Log and Report
  • Log the review outcome in logs/claude_log.md.
  • Report verdict and finding count to the user.

Output

  • GitHub PR review posted with structured findings.
  • Each finding has: file, line, severity, description, suggestion.
  • Summary verdict with rationale.

Review Checklist

  • [ ] All changed files examined in the diff.
  • [ ] Security scan completed — no secrets or injection risks.
  • [ ] Performance scan completed — no N+1 or blocking operations.
  • [ ] Style scan completed — conventions followed.
  • [ ] Findings consolidated with correct severity ratings.
  • [ ] Review posted to GitHub via gh CLI.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
1
7-day trend score
8

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/knofler/myai/github-pr-review/badge)](https://verified-skill.com/skills/knofler/myai/github-pr-review)

Something wrong with this skill or its page? Tell us.