This is a comprehensive AI-powered code review assistant that orchestrates multiple static analysis tools (CodeQL, SonarQube, Semgrep) alongside LLMs like Claude and GPT-4 to catch bugs, security flaws, and architectural issues in pull requests. It routes reviews based on complexity, runs parallel scans for OWASP vulnerabilities and secret leaks, and generates line-specific feedback with fix examples. The routing logic is smart: small PRs get fast models, security-sensitive changes get deeper reasoning, and anything over 1000 lines escalates to humans. Built for CI/CD integration with support for 30+ languages. The playbook is thorough but borders on over-engineered for teams that just need quick feedback. Best for organizations serious about automated code quality gates.
npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill performance-testing-review-ai-review