Meta-cognitive framing for analyze-before-doing, ownership routing, risk gating, minimum-closure planning, and retrospective extraction in multi-agent work. Use when the user says 先分析再做 / 先别动手 / 先判断 / 先定方案, when work spans multiple agents or needs dispatch/orchestration, when CEO-style delegation or group command requires owner+deadline+closure, when abnormal sessions/cron/jobs/runs need real follow-through instead of status-only reporting, or when a task needs strategy, PRD, first-principles thinking, verification, and postmortem/retro.
Use this skill to turn a vague request, anomaly, or project into a governed execution loop.
This skill is for judgment before action and closure after action. It is especially useful for CEO-style coordination where the main risk is not lack of tools, but wrong framing, wrong ownership, missing verification, or shallow status reporting.
For any non-trivial task, produce these six sections before major execution:
If the user asks for a fast answer, compress the six sections into short bullets instead of skipping them.
Use full strict mode when any of the following is true:
In strict mode, do not jump from detection straight to execution. Frame → route → gate → act → verify.
Rewrite the request into the real problem.
Answer:
Rules:
Decide who should do the work.
Answer:
Rules:
For detailed routing heuristics, read references/ownership-routing.md.
Assign a risk level before acting.
Use three levels:
For every task, state:
Rules:
Define the smallest end-to-end result that counts as actually finished.
Examples:
State:
For closure patterns, read references/closure-loop.md.
Only now decide what to do.
Use one of four modes:
When dispatching:
If useful, use this structure:
Before saying “done”, check:
Never collapse “looks good” into “done”.
If the task produced a reusable lesson, explicitly propose one of:
For retro extraction prompts, read references/retro-prompts.md.
Use this template unless the user asks for a different format:
## 1. 问题本质
- 表层需求：
- 真问题：
- 成功标准：
- 未知项：
## 2. 责任归属
- 主负责：
- 协同：
- CEO 是否亲自执行：
- 是否并行：
## 3. 风险等级
- Level：P0 / P1 / P2
- 失败模式：
- 是否需确认：
## 4. 最小闭环
- 交付物：
- 验证方式：
- 完成证据：
- 当前不做：
## 5. 执行动作
- 模式：Act / Dispatch / Ask / Defer
- 下一步：
## 6. 验证与复盘
- 已验证：
- 未验证：
- 可沉淀项：
Do not do these:
This skill is a strong match for prompts like:
Read bundled references only when needed:
references/ownership-routing.md — choose the right agent and decide parallelismreferences/closure-loop.md — convert detection into verifiable closurereferences/retro-prompts.md — extract durable lessons without noise