基于 SOC 职业分类
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。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/Archive228/loopkit --skill spec-first命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
Before compaction Loopkit extracts decisions into claude-decisions.json (machine-readable). Read it alongside claude-progress.txt at session start — prose is for humans, JSON is for the loop.
Split the Plan/Act/Verify loop across three model tiers — frontier planner, cheap executor, frontier judge — via env vars read by run.sh.
Run the fixed 6-step session-opening sequence — pwd, read progress, git log, count remaining features, init.sh, smoke-test last feature — before touching any new work. The orientation ritual that lets fresh-context sessions reconstruct project state in under a minute.
| name | spec-first |
| description | Write the goal spec on disk before the agent acts, so it can't drift. Use before any multi-step task. |
| when_to_use | a task with more than 2 steps, a long-running job, "build X", agent drifting |
Without an external contract, the agent drifts after ~3 iterations — and the failure looks like progress (code written, tests pass, wrong goal solved).
Write PROMPT.md BEFORE acting:
IMPLEMENTATION_PLAN.md it updates in place. If you can't write "done when" concretely, the task isn't ready — clarify before coding.