task-decomposer
Skill1 starsChanged 3 months agoMITView on GitHub
任务拆解。当面对一个大功能、大目标或多个步骤组成的复杂任务时,把任务拆成依赖清晰、可独立完成、带验收标准的子任务清单。Use to decompose large features or goals into small, verifiable, dependency-ordered subtasks.
9 other projects use a version of this.
Adopted by
Most recently taken up by
- Mathews-Tom/armorytheir own version, first seen Nov 1, 2026
- nizaalkhot/skillstheir own version, first seen Nov 1, 2026
- fatih-developer/fth-skillstheir own version, first seen Oct 13, 2026
- yasdelayu/agent-x-cctheir own version, first seen Oct 13, 2026
- Alteriom/ai-dev-skillstheir own version, first seen Oct 11, 2026
- GhostwheeI/Master-Plannertheir own version, first seen Jun 16, 2026
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 9
- Mathews-Tom/armory★ 340same name
- ttmouse/skills★ 34same name
- gear-foundation/vara-skills★ 17same name
- fatih-developer/fth-skills★ 7same name
- Alteriom/ai-dev-skills★ 0same name
- dhriyatinandu-bot/claude-skills★ 0same name
- GhostwheeI/Master-Planner★ 0same name
- nizaalkhot/skills★ 0same name
- yasdelayu/agent-x-cc★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository28 other files it gives its agents.
- api-contract-designerSkill
- api-testerSkill
- bug-diagnoserSkill
- cache-governorSkill
- ci-cd-setupSkill
- code-migratorSkill
- code-reviewerSkill
- commit-message-writerSkill
- db-migration-reviewerSkill
- db-schema-designerSkill
- dev-env-troubleshooterSkill
- frontend-debugSkill
- git-workflowSkill
- log-analysisSkill
- meeting-minutesSkill
- natural-chinese-writerSkill
- performance-profilerSkill
- release-note-writerSkill
- requirement-clarifierSkill
- requirements-reviewerSkill
- security-reviewerSkill
- skill-creator-cnSkill
- skill-style-guideSkill
- sql-reviewerSkill
- tdd-workflowSkill
- tech-design-writerSkill
- tech-doc-writerSkill
- weekly-reportSkill
The file
Files
任务拆解
何时使用
- 用户说"把这个功能做完"但没说从哪开始;
- 任务包含多个步骤,需要排执行顺序;
- 需要把大任务变成可估算、可验收的小任务;
- 开发前规划 commit 粒度,让每个 commit 对应一个子任务。
使用步骤
第 1 步:明确目标与边界
确认三件事:
- 最终交付物是什么(代码、文档、部署?);
- 范围与不做项;
- 约束(时间、技术栈、依赖)。
第 2 步:自顶向下拆解
按"功能模块 → 子功能 → 单任务"逐层拆解,直到每个任务满足:
- 单一职责:只做一件事;
- 可独立完成:不依赖未完成的他人任务;
- 可验收:有明确的完成标准;
- 可估算:能估计工作量(如 0.5 天、2 小时);
- 可提交:完成后能形成一次有意义的 commit。
第 3 步:确定依赖与顺序
用依赖关系排出执行顺序:
任务 A(定义数据模型)
↓
任务 B(编写 API 接口)→ 任务 C(前端页面)
↓
任务 D(联调与测试)→ 任务 E(部署)可并行的任务标注"可并行"。
第 4 步:输出任务清单
模板、判断标准和常见拆法见 references/DECOMPOSE_TEMPLATE.md。
# 任务清单:<目标>
| # | 任务 | 依赖 | 交付物 | 验收标准 | 预估 |
|---|------|------|--------|----------|------|
| 1 | 定义数据模型 | - | schema 文件 | 覆盖全部字段与关系 | 2h |
| 2 | 编写创建接口 | 1 | API + 测试 | 单测通过,文档更新 | 4h |
| 3 | 实现前端表单 | 2 | 页面组件 | 能提交并看到结果 | 4h |
## 推荐提交顺序
1 → 2 → 3(每个任务一个 commit)写完用 task_checker.py 自查(见下面的辅助脚本),修掉标出的缺口再交付。
输入与输出
- 输入:一个目标或大任务描述;
- 输出:带依赖、交付物、验收标准的任务清单。
示例
用户输入: "做一个用户注册登录功能。"
拆解结果:
| # | 任务 | 依赖 | 交付物 | 验收标准 |
|---|------|------|--------|----------|
| 1 | 用户表设计 | - | 迁移文件 | 含 username、password_hash、created_at |
| 2 | 注册接口 | 1 | API + 测试 | 重复用户名返回 409,密码哈希存储 |
| 3 | 登录接口 | 1 | API + 测试 | 正确密码返回 token,错误密码 401 |
| 4 | 前端注册/登录页 | 2,3 | 页面 | 注册成功跳转登录,错误有提示 |
| 5 | 会话保持 | 4 | 中间件 | 刷新页面不退出登录 |辅助脚本
scripts/task_checker.py 检查任务清单表:必需列是否齐全、任务编号是否重复、依赖是否存在且无环、验收标准是否为空或含糊、有无占位符:
python skills/task-decomposer/scripts/task_checker.py <任务清单.md>脚本只做标记,结论需要人确认。模板见 references/DECOMPOSE_TEMPLATE.md。
注意事项
- 拆到可执行为止:如果用户说"不知道怎么开始",就拆到第一个任务可以直接动手;
- 保留弹性:标注不确定的任务为"需调研",不要假装精确;
- 不要过度拆解:单任务小到不需要验收标准就是过度拆解;
- 配合 requirement-clarifier 使用:先澄清需求,再拆任务。
不适用场景
- 任务本身很小、一步可完成;
- 用户只需要粗略的步骤列表,不需要验收标准;
- 目标是开放性研究(如"调研一下"),建议先用调研型技能。
验证方式
- 触发:"把这个功能拆成任务""帮我规划一下实现步骤";
- 跑脚本:
python skills/task-decomposer/scripts/task_checker.py <任务清单.md>,无"必须"级问题; - 检查:每个任务满足"单一职责、可验收、有依赖说明";
- 走查:按清单顺序执行第一个任务,确认没有前置缺口。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

