Activates when planning, code review, or any analytical task requires multi-perspective critical examination. Guides internal reasoning to identify hidden assumptions, logical gaps, and overlooked risks. Do NOT use for simple Q&A. Examples: 'Review this design plan', 'Evaluate this approach'.
A Skill is a reusable instruction package (usually a SKILL.md file) that teaches an AI agent how to complete a specific task with the right context, tools, and steps.
Agent Skills are packaged capabilities that agents such as Claude Code and Codex can discover, load, and execute on demand.
This skill is designed for agent runtimes that support SKILL.md workflows, including OpenClaw, Claude Code, Codex, and similar AI coding agents.
Open the source repository and follow its installation notes. Review any scripts and permissions before enabling it.
Once installed, ask your agent to use the skill by name or let the agent discover it from the skill directory.
Yes. Skills are intended to be adapted, but review the license and keep safety-critical instructions intact.
Share practical notes and help others evaluate this skill.
No comments yet. Start the conversation.
