bootstrap

Create new agent skills from scratch, improve or optimize existing skills, and measure skill performance with evaluations. Use when the user wants to create a skill, edit or improve an existing skill, run evals or benchmarks to test skill quality, optimize a skill's description for better triggering, or package a skill for distribution. Works across Claude Code, OpenAI Codex, GitHub Copilot, OpenCode, Gemini, and other agents supporting the Agent Skills open standard.

Analyse code for complexity issues and suggest reductions. Use when the user asks to check complexity, simplify code, or reduce technical debt.

Review code changes for correctness, security, clarity, and adherence to project conventions. Use when the user asks to review code, check a diff, or audit a PR.