Github Code Review

This deploys multiple specialized AI agents to review your PRs in parallel, each focused on a different concern like security, performance, or architecture. You feed it PR data from the GitHub CLI and it coordinates a swarm of reviewers that can block on critical security issues, suggest optimizations, or auto-fix style violations. The label-based agent assignment is smart: tag something "security" and it automatically spawns authentication and audit agents. Best for teams that want deeper code review than linters provide but don't have bandwidth for manual deep dives on every PR. The multi-agent approach means you get security, performance, and architecture feedback simultaneously instead of sequentially.

Install to Claude Code

npx -y skills add ruvnet/ruflo --skill github-code-review --agent claude-code