菜单
Use when you have a written implementation plan to execute in a separate session with review checkpoints
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Orchestrated multi-agent code review modeled on Cloudflare's system. Risk-tiers the diff, filters noise, fans out specialist reviewers (security, performance, code quality, docs, tests, release, AGENTS.md), and consolidates everything through a coordinator judge pass into one structured verdict. Use for "orchestrated review", "multi-agent code review", "deep review", "review like Cloudflare", "thorough AI code review", or to review a GitHub PR with severity-graded findings and an approve/block decision.
Document a recently solved problem into a searchable project-local knowledge store so each unit of engineering work makes the next one easier. Use after a non-trivial problem is solved and verified, when the user says "that worked" / "it's fixed" / "problem solved", or when invoked headless by the lfg pipeline. Writes one doc to docs/solutions/ and makes it discoverable.
Use when executing implementation plans with independent tasks in the current session
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
Use when you have a spec or requirements for a multi-step task, before touching code
Run the full autonomous engineering pipeline end-to-end (recall prior learnings, plan, implement with TDD, review and autofix, verify, commit, push, open PR, watch CI and fix failures, then compound the learning). Use ONLY when the user explicitly requests hands-off execution of a software task and provides a feature description. Do not auto-route casual conversation here.
| name | executing-plans |
| description | Use when you have a written implementation plan to execute in a separate session with review checkpoints |
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Note: Tell your human partner that Superpowers works much better with access to subagents. The quality of its work will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use references/subagent-driven-development/SKILL.md instead of this skill.
For each task:
After all tasks complete and verified:
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
Required workflow skills:
