菜单
Use when the input is a complex task description that needs decomposition, persistence, and recovery before execution through a generated .mission CSV.
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
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.
Use when executing an existing task CSV and the agent needs to push all actionable rows to closed-loop completion without stopping between issues.
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.
Use when a task may be a user-approved design/plan document, an existing task CSV, a markdown execution or investigation document, a long-running task description, or a recovery request and needs routing to the right mission sub-skill.
Use when the input is a markdown document and the agent must decide whether it belongs on the approved-doc execution path or the long-task execution path.
| name | mission-long-task |
| description | Use when the input is a complex task description that needs decomposition, persistence, and recovery before execution through a generated .mission CSV. |
你现在是「长任务分解与执行器」。
把一个复杂任务分解为可追踪的 .mission/<stem>/<stem>.csv，然后交给 mission-csv-execute 闭环执行。
<short-topic>（小写，连字符分隔），再生成 run stem：<YYYYMMDD_HH-mm-ss>-<short-topic>.mission/ 目录存在先运行 superpowers:systematic-debugging，捕获根因分析，用发现指导任务拆分。
.mission/<stem>/<stem>.csv，使用标准 CSV schemaid, priority, phase, area, title, description, acceptance_criteria, test_mcp, required_skills, required_mcpREVIEW-01，用于按 mission-csv-execute 的 review capability 阶梯做长任务完成度 review；该行必须包含从原始任务抽取的任务专属 claim/evidence 检查项，不能只写通用套话acceptance_criteria 必须可机器验证或有明确复现步骤refs 至少 1 个 path:linerequired_skills 与 required_mcp.mission REVIEW-01 行规则.mission/<stem>/<stem>.csv 也必须显式生成 review 行；不要只依赖最后总结。Legacy .mission/*.csv 只为恢复兼容。
生成 REVIEW-01 时，先从原始任务中抽取用户真正承诺的结果，并写进 review 条件：
validation_limited / manual_test / risk，不得用替代假路径伪装通过review_regression_requirements 必须包含 2-4 条来自原始任务的任务专属检查项；如果无法抽取，至少写明要审查 claim/evidence 是否一致建议字段：
| 字段 | 值 |
|---|---|
id | REVIEW-01 |
priority | P0 |
phase | 最后阶段序号 |
area | review |
title | Review task outcome against original request |
description | Run the strongest available independent vision review to compare original-request claims with delivered behavior, evidence level, CSV state, validation evidence, and mission log. |
acceptance_criteria | WHEN all non-review issues before this row are closed THEN run vision review through mission-csv-execute's review capability ladder against source-specific claim/evidence checks; WHEN gaps or overstated claims are found THEN append follow-up issues and REVIEW-02; WHEN no gaps remain THEN close the CSV. |
test_mcp | MANUAL |
required_skills | superpowers:requesting-code-review |
required_mcp | 留空，除非任务本身要求浏览器或外部验证 |
review_initial_requirements | Verify all prior non-review rows are closed before running this review. |
review_regression_requirements | Run strongest available independent vision review against original request, acceptance criteria, delivered diff, validation evidence, mission log, and source-specific claim/evidence alignment checks. |
refs | <best-source-path-or-request>:1 |
notes | review_kind:vision; review_agent_mode:pending; review_independence:pending; source_doc:<task-source> |
将生成的 .mission/<stem>/<stem>.csv 交给 mission-csv-execute。
mission-csv-execute，不要因为 CSV 已生成、log 已齐、checkpoint 已完整而暂停。与批准文档执行流的区别：
.mission/<stem>/，作为本地恢复工件[<task-name>-<id>] <title>被中断后恢复时：
.mission/*/*.csv，再兼容 legacy .mission/*.csvlog.md（如有）→ 恢复决策上下文上下文已恢复
任务: <from CSV>
进度: X/Y 步已完成
恢复点: Step #N - <title>
对于长任务，维护 CSV 同目录的 log.md：
## Step N: <title>
- **状态**: DONE | FAILED
- **做了什么**: ...
- **关键决策**: ...
- **遇到的问题**: ...
- **变更文件**: path:line
- **下一步**: Step N+1
log.md 是恢复工件，不是自然停点。记录完成后继续执行，除非已满足 mission-csv-execute 的正式停止条件。log.md 写最终总结（如有）.mission/ # gitignored，仅辅助工件
└── <stem>/
├── <stem>.csv # 任务状态（标准 CSV schema）
├── <stem>.review.md # review log（如有）
├── <stem>.handoff.md # human handoff（如有）
├── log.md # 决策日志 / 审计轨迹
├── reviews/ # reviewer 原始输出 / prompt / jsonl（如有）
└── raw/ # 缓存的外部数据