菜单
How to know when you're done. Run tests, check the spec. If both pass, stop. If the same issue persists after 3 iterations, write a failure note and stop.
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
Complete reference for the factory CLI — all commands, arguments, options, and usage. Use when an agent needs to invoke factory commands, check correct syntax, or understand available options.
When and how to write a human-action-needed entry to memory/{agent}/needs.md. Agents write these entries when they hit blockers that only a human can resolve.
When you encounter a failure, correction, or new pattern, write a learning and optionally propose a skill.
How the factory workspace is organized and what each directory means.
What to examine during a reflection pass and how to write observations to memory/{agent}/needs.md.
When and how to write scenario holdouts for factory verification.
| name | convergence |
| description | How to know when you're done. Run tests, check the spec. If both pass, stop. If the same issue persists after 3 iterations, write a failure note and stop. |
After each implementation attempt, run this check in order:
If both (1) and (2) are satisfied, stop. Do not add features the spec doesn't mention.
Track the number of times you have attempted to fix the same failing criterion.
A failure note goes to tasks/review/{task-name}.md with a failure: true flag and
an explanation of what you tried, what the spec requires, and why you could not
converge. Do not write to tasks/review/ without first writing this note — it is
not Builder's job to silently deliver broken work.
Done does not mean:
