A code review graph builds a map of your entire codebase. It tracks how files, functions, and classes connect to each other. This map is stored locally on your machine.
When you make a change, the graph finds the blast radius of that change. It discovers every file that might be affected. An AI like Claude then reads only those files instead of the whole project.
This means fewer tokens are used per review. In large projects you can save up to 49 times the usual amount. Reviews become faster and cheaper.
Global
mkdir -p ~/.claude/skills/code-review-graphProject
mkdir -p .claude/skills/code-review-graphSource Repository
Web Design Guidelinesvercel-labs/agent-skills
Review your UI code for compliance with web interface guidelines
Improve Codebase Architecturemattpocock/skills
Scan your codebase for architecture improvements with visual reports and pick one to refactor
Zoom Outmattpocock/skills
Get a big picture view of your code with a simple map of modules and callers
Caveman Reviewjuliusbrussee/caveman
Ultra-compressed one-line code review comments that cut noise
Requesting Code Reviewobra/superpowers
Catch code issues early with focused automated reviews before every merge
Receiving Code Reviewobra/superpowers
Think first, then code. Verify feedback before making changes. Stay objective
Auditpbakaus/impeccable
Audit your code for accessibility performance theming and anti-patterns
Request Refactor Planmattpocock/skills
Plan safe code refactors with tiny commits and clear GitHub issues