This is the code review you run before you ship. Point it at your markdown specs or PRD and it'll trace every requirement to actual implementation, flag what's missing, what's half-done, and what you overbuilt. It runs a methodical nine-pass audit covering correctness, integration, security, performance, and cleanup. You can use it in review-only mode to get findings with file and line evidence, or ask it to fix and it'll make targeted changes without scope creep. The workflow is thorough without being precious: it reads your docs, inspects the relevant surfaces, builds a traceability map, then either reports or repairs. Honest take: if you work from written specs, this closes the loop between what you said you'd build and what you actually built.
npx -y skills add golbin/agent-skills --skill review-implementation --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
github/awesome-copilot
microsoft/win-dev-skills
github/awesome-copilot
github/awesome-copilot