Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Review Forge structures code review across multiple LLMs or agents: each model produces a separate review file, then the skill synthesizes them into a human-editable checklist. Only checked items are fixed, and every fix batch triggers tests and an independent verification pass. It keeps review artifacts like `summary.md`, `fix-plan.md`, and `status.md` isolated from product code. Teams working with multiple code review models get a repeatable workflow that separates gathering review evidence from acting on it.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Installs: Complete review-forge skill installed inside one project
Before you start
Run from: Target project root
From the target project's root directory, run this complete terminal command block to create the local skills directory and clone review-forge into it.
Claude
Codex
Cursor
Gemini CLI
OpenCode