Improve AI application with evaluation-driven development. Define eval criteria, instrument the application, build golden datasets, observe and evaluate application runs, analyze results, and produce a concrete action plan for improvements.
Install
npx skills add github/awesome-copilot --skill eval-driven-devnpx skills add github/awesome-copilot --skill eval-driven-dev -a claude-code -gnpx skills add github/awesome-copilot --skill eval-driven-dev -a codex -ggemini skills install https://github.com/github/awesome-copilot.git --path skills/eval-driven-devgh skill install github/awesome-copilot eval-driven-devgit clone --depth 1 https://github.com/github/awesome-copilot /tmp/github__awesome-copilot
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/github__awesome-copilot/skills/eval-driven-dev ~/.claude/skills/eval-driven-dev # Claude Code
cp -R /tmp/github__awesome-copilot/skills/eval-driven-dev ~/.agents/skills/eval-driven-dev # Codex, Cursor, Gemini CLI, Copilot, OpenCode