Summary
This is a structured PR review tool that reads diff and description files from your local checkout, analyzes the changes, and writes machine-readable feedback to review.json instead of posting directly to GitHub. It's built for CI workflows where you want Claude to review code but need the output in a specific JSON shape for another step to handle publishing. The prompt is opinionated about what matters: it prioritizes correctness and security, avoids nitpicking unless it can offer concrete fixes, and labels every comment as critical, important, suggestion, or nit. The diff annotation format with OLD/NEW line markers is smart but adds learning overhead. If you're wiring Claude into a review pipeline and need deterministic output rather than conversational feedback, this gives you the guardrails.
Install to Claude Code
npx -y skills add warpdotdev/common-skills --skill review-pr --agent claude-codeInstalls into .claude/skills of the current project.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Files
SKILL.md
Select a file.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
