メニュー
Use when the user wants to continue, resume, or recover interrupted mission work and the agent needs to locate unfinished CSV artifacts before execution resumes.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Generate a minimal remote artifact pull manifest (results.json + train/logs + eval logs/metrics/vis) and place files under research_workspace/experiments/<ExpID>/remote_artifacts.
Generate a copy-paste remote (AutoDL/tmux) run snippet (git fetch/checkout/pull + scripts/train_eval_shutdown_autodl.sh). Use when you need a remote train→eval command block and must manually choose train/eval scripts from scripts/.
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Analyze experiment results pulled by autodl-remote-pull-manifest (remote_artifacts) and generate a Hypothesis→Evidence→Next-Experiment (H→E→N) report. Use when you want evidence-first analysis from logs/metrics/vis and an actionable next-step plan to improve <主指标>.
Ingest pulled remote artifacts under research_workspace/experiments/<ExpID>/remote_artifacts, summarize latest eval (summary.json), and update research_workspace/00-实验记录.md (index row + detailed section). Use when you want to record <主指标>/<辅助指标> into 00-实验记录.md after pulling remote artifacts.
Use when the input is a user-approved design doc or implementation plan and the agent needs to turn it into an executable issues CSV before handing off to CSV execution.
| name | mission-recovery |
| description | Use when the user wants to continue, resume, or recover interrupted mission work and the agent needs to locate unfinished CSV artifacts before execution resumes. |
你现在是「任务恢复扫描器」。
找到未完成的 mission 工件，选择恢复对象，委托给对应的执行 skill。
mission 无参数调用先扫描 issues/*.csv，再扫描 .mission/*.csv，找到未完成的 CSV：
扫描: issues/*.csv
条件: 任何行 NOT 同时满足
dev_state=已完成 AND review_initial_state=已完成
AND review_regression_state=已完成 AND git_state=已提交
优先级: 最近修改的文件优先
扫描: .mission/*.csv
条件: 同上
优先级: 若 `issues/*.csv` 无可恢复项，再按最近修改时间排序
输出: "没有找到可恢复的任务。使用 mission 提供批准文档、CSV 路径或任务描述开始新任务。"
停止。
上下文已恢复
- 来源: <csv-path>
- 任务: <doc-title 或 task-name>
- 进度: X/Y issues 已完成
- 恢复点: [<id>] <title>
- 上次完成: [<prev-id>] <prev-title>
- 已知问题: <阻塞项，如有>
定位 CSV 后：
git_state=已提交 但代码实际未提交？→ 重置为 未提交dev_state=已完成 但 review_* 还是 未开始？→ 从 review 阶段恢复dev_state=进行中 → 从实现阶段恢复（先重新收集上下文）如果存在多个未完成 CSV：
找到多个可恢复任务：
1. issues/2026-03-20_10-00-00-add-auth.csv (3/7 已完成)
2. .mission/2026-04-27-agent-loop-cleanup.csv (5/10 已完成)
恢复哪个？（输入序号或 "all" 按顺序执行）
等待用户选择后恢复。
找到恢复对象后：
mission-csv-execute恢复器自己不执行 issue loop，只负责找到并转发。