cs
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinesecs
cs
cscs-xxxcs两件事,仅此两件:
- 当用户带着具体诉求来 → 匹配场景路由表,告诉用户该触发哪个 ,并简单说明为什么是这个
cs-* - 当用户只是想了解体系 / 说不清想做什么 → 给一段精简的体系速读 + 让用户挑或描述更具体的诉求
本技能不做事:不写 spec、不读写 下的内容产物、不替子技能跑流程。它的产出只有"建议触发哪个子技能"。
codestable/cscs-xxxcsOnly two tasks, nothing more:
- When users come with specific requests → match the scenario routing table, tell users which to trigger, and briefly explain why
cs-* - When users just want to learn about the system / can't clearly state what they want to do → provide a concise system overview + let users choose or describe more specific requests
This skill does not perform actual tasks: it doesn't write specs, read/write content products under , or run processes on behalf of sub-skills. Its only output is "suggest which sub-skill to trigger".
codestable/收到调用先做的扫描
Scans to perform first upon receiving a call
在回应用户之前先做这些(每次都做,不长,几个 tool 调用就够):
- 看仓库有没有接入 CodeStable——看顶层目录是否存在
Glob codestable/ - 如果存在——(如果有)作为更全的体系参考;
Read codestable/reference/system-overview.md一下Globcodestable/features/codestable/issues/看现在进行中的工作有哪些(拿目录名就够,不要逐份读)codestable/roadmap/ - 如果不存在——记下来,后面提示用户先走
cs-onboard - 看用户原话——TA 是开放式提问还是带着具体诉求?带诉求就匹配路由表,没诉求就给体系介绍
扫完才能回应。让用户感觉你心里有数,不是凭空在问。
Do these before responding to users (do this every time; it's quick, just a few tool calls):
- Check if the repository has integrated CodeStable — use to see if the top-level directory exists
Glob codestable/ - If it exists — read (if available) as a more comprehensive system reference; use
codestable/reference/system-overview.mdonGlobcodestable/features/codestable/issues/to see ongoing work (just get the directory names, no need to read each file)codestable/roadmap/ - If it does not exist — note this, and later prompt users to run first
cs-onboard - Check the user's original input — is it an open-ended question or a specific request? If it's a request, match the routing table; if not, provide the system overview
Respond only after completing these scans. Make users feel you have a clear understanding instead of responding blindly.
体系一图速读(用户没具体诉求 / 让你介绍时讲这个)
Quick system overview (explain this when users have no specific requests / ask for an introduction)
CodeStable 把软件开发活动建模成 6 个实体 + 3 个流程,所有产物都聚在项目根的 目录下:
codestable/codestable/
├── requirements/ 需求实体("为什么要有这个能力",只记现状)
├── architecture/ 架构实体("系统现在长什么样",只记现状)
├── roadmap/ 规划层("接下来打算怎么做这块大需求 + 模块怎么切 + 接口怎么定")
├── features/ 新增能力的 spec 聚合根(design / impl / accept)
├── issues/ 修 bug 的 spec 聚合根(report / analyze / fix)
├── refactors/ 重构的 spec 聚合根(beta)
└── compound/ 知识沉淀(learning / trick / decision / explore)三条流程:
- 新增能力:→
cs-feat-design→cs-feat-impl(想法模糊先走cs-feat-accept分诊)cs-brainstorm - 修 bug:→
cs-issue-report→cs-issue-analyzecs-issue-fix - 重构(beta):/
cs-refactorcs-refactor-ff
横切:任何流程跑完发现"这事值得记下来" → / / / 沉淀到 。
cs-learncs-trickcs-decidecs-explorecompound/核心理念:编排的是软件本身的生命周期(需求、架构、特性、bug、决策),而不是 Agent。人在环——程序员对整体把控负责,AI 是高效的执行体。
项目已 onboard 的话,更详细的总览看。codestable/reference/system-overview.md
CodeStable models software development activities into 6 entities + 3 processes, with all products gathered in the directory at the project root:
codestable/codestable/
├── requirements/ Requirement entity ("why this capability is needed", only records current status)
├── architecture/ Architecture entity ("what the system looks like now", only records current status)
├── roadmap/ Planning layer ("how to implement this large requirement next + module splitting + interface definition")
├── features/ Aggregation root for specs of new capabilities (design / impl / accept)
├── issues/ Aggregation root for specs of bug fixes (report / analyze / fix)
├── refactors/ Aggregation root for refactoring specs (beta)
└── compound/ Knowledge accumulation (learning / trick / decision / explore)Three processes:
- New capability development: →
cs-feat-design→cs-feat-impl(start withcs-feat-accepttriage if ideas are vague)cs-brainstorm - Bug fixing: →
cs-issue-report→cs-issue-analyzecs-issue-fix - Refactoring (beta): /
cs-refactorcs-refactor-ff
Cross-cutting: After any process, if "this is worth recording" → use / / / to store in .
cs-learncs-trickcs-decidecs-explorecompound/Core concept: It orchestrates the lifecycle of the software itself (requirements, architecture, features, bugs, decisions), not Agents. Human-in-the-loop — programmers are responsible for overall control, and AI is an efficient executor.
If the project has been onboarded, seefor a more detailed overview.codestable/reference/system-overview.md
场景路由表
Scenario routing table
匹配用户的话到表里某行,然后告诉用户:"你这个诉求建议走 ,因为 {一句话理由}"。
cs-xxx| 用户说什么 / 想做什么 | 路由到 |
|---|---|
仓库还没有 | 先 |
| 想法还模糊 / "有个想法没想清楚" / "先聊聊" / "不知道是不是新功能" | |
| 新功能 / 新能力 / "加个 X" / "实现 XX" | |
| BUG / 异常 / 报错 / "这里不对" / "文档错了" | |
| 代码优化 / 重构 / 重写(行为不变) | |
| 摸代码 / "X 是怎么实现的" / 提问调研 | |
| 补 / 更新需求文档 | |
| 补 / 更新 / 检查架构文档 / "刷新架构 doc" / "做架构体检" | |
| 大需求拆解 / "我想要一个 X 系统" / 排期规划 / 设计模块拆分 + 接口契约 | |
| 技术选型 / 长期约束 / 编码规约 | |
| 踩坑回顾 / 经验总结 / "这事值得记下来" | |
| 可复用编程模式 / 库用法 / "以后做 X 就该这样" | |
| 开发者指南 / 用户指南 | |
| 库 API 参考 | |
| 用户在 feature / issue 流程中间,问"下一步该干啥" | 路由到对应流程的入口( |
判不出来 / 用户说得太抽象:"听起来像 {猜测},但你描述里 {缺什么}。是 {选项 A} 还是 {选项 B}?" 让用户选,不要硬猜。
Match the user's input to a row in the table, then tell the user: "For this request, it is recommended to use because {one-sentence reason}".
cs-xxx| User's input / What the user wants to do | Route to |
|---|---|
No | First run |
| Vague ideas / "I have an idea but haven't figured it out" / "Let's chat first" / "Not sure if it's a new feature" | |
| New feature / new capability / "Add X" / "Implement XX" | |
| BUG / exception / error / "This is wrong" / "Documentation is incorrect" | |
| Code optimization / refactoring / rewrite (behavior unchanged) | |
| Explore code / "How is X implemented" / research questions | |
| Supplement / update requirement documentation | |
| Supplement / update / check architecture documentation / "Refresh architecture doc" / "Do architecture checkup" | |
| Large requirement breakdown / "I want an X system" / scheduling planning / design module splitting + interface contracts | |
| Technology selection / long-term constraints / coding conventions | |
| Pitfall review / experience summary / "This is worth recording" | |
| Reusable programming patterns / library usage / "Do X like this in the future" | |
| Developer guide / user guide | |
| Library API reference | |
| User is in the middle of a feature / issue process and asks "What to do next" | Route to the entry of the corresponding process ( |
If unable to match / user's input is too abstract: "It sounds like {guess}, but {what's missing} in your description. Is it {Option A} or {Option B}?" Let the user choose, don't guess blindly.
几种需要特别留心的情况
Special cases to pay attention to
仓库还没接入
Repository not integrated yet
用户开口想做任何 cs-* 流程的事,但 不存在 → 说明这一点,建议先 。不要自作主张直接路由到 cs-feat / cs-issue 等——它们的 SKILL.md 都假设 已经存在。
codestable/cs-onboardcodestable/If the user wants to perform any cs-* process but does not exist → explain this and recommend running first. Do not route directly to cs-feat / cs-issue etc. on your own — their SKILL.md all assume already exists.
codestable/cs-onboardcodestable/大需求被误当成 feature
Large requirements mistaken as features
用户说的是"我想要一个权限系统 / 通知中心 / SSO 接入"这类一眼看出做不完一个 feature 的诉求 → 不要路由到 ,路由到 (大概率会判为 case 3 → )或直接 。理由:直接起 feature 会变成巨型 design 塞不下。
cs-featcs-brainstormcs-roadmapcs-roadmapIf the user mentions requests like "I want a permission system / notification center / SSO integration" which are obviously too big to complete as a single feature → do not route to , route to (most likely will be judged as case 3 → ) or directly to . Reason: Starting a feature directly will result in an oversized design that can't be accommodated.
cs-featcs-brainstormcs-roadmapcs-roadmap"改一下 X" 但 X 是已有功能
"Modify X" but X is an existing feature
用户说"改一下 X 的 Y 行为" → 先问一句这是 bug 修复(X 现在表现错了)还是 需求变更(X 现在的表现没错,但产品策略变了):
- bug →
cs-issue - 需求变更 → (改需求文档)+ 之后
cs-req跑实现cs-feat
If the user says "Modify the Y behavior of X" → first ask whether this is a bug fix (X's current behavior is wrong) or a requirement change (X's current behavior is correct, but product strategy has changed):
- Bug →
cs-issue - Requirement change → (update requirement documentation) + then run
cs-reqfor implementationcs-feat
进行中的工作
Ongoing work
扫描看到 或 下已经有跟用户描述相关的目录 → 提一句"看到 已经存在,是接着做这个吗?" 让用户确认是续作还是开新的,避免无意中创建重复 feature。
codestable/features/codestable/issues/features/2026-04-22-xxx/If scanning finds directories related to the user's description under or → mention "Noticed already exists, do you want to continue working on this?" Let the user confirm whether to continue or start a new one, to avoid accidentally creating duplicate features.
codestable/features/codestable/issues/features/2026-04-22-xxx/沉淀类技能的细分
Subdivision of accumulation skills
learning / trick / decision / explore 容易混。判别口诀:
- 回顾"做 X 时踩了 Y" →
cs-learn - 处方"以后做 X 就这样做" →
cs-trick - 规定"全项目今后都按 X 来" →
cs-decide - 调查"X 现在是什么样" →
cs-explore
判不出问用户:"这个你想记成 {踩坑回顾 / 复用处方 / 长期规约 / 调研存档} 哪一种?"
learning / trick / decision / explore are easily confused. Discrimination mnemonic:
- Review "Encountered Y when doing X" →
cs-learn - Prescription "Do X like this in the future" →
cs-trick - Rule "All projects follow X from now on" →
cs-decide - Investigation "What is X like now" →
cs-explore
If unable to judge, ask the user: "Do you want to record this as {pitfall review / reusable prescription / long-term convention / research archive}?"
介绍模式(用户只说想了解 / 不知道做什么)
Introduction mode (when users only want to learn / don't know what to do)
按这个顺序讲,不要一次倒出全部:
- 一句话:CodeStable 是面向严肃工程的 AI 编码工作流,编排软件生命周期而不是 Agent
- 6 实体 + 3 流程的速读图(见上面"体系一图速读")
- 问用户:"你现在最想从哪儿开始?" 给三个具体引子:
- "我有个新功能想做" → cs-feat
- "代码里有个 bug" → cs-issue
- "项目还没接入 CodeStable" → cs-onboard
收住,别把所有子技能的细节都讲一遍。用户问到具体某个再展开。
Explain in this order, do not dump all information at once:
- One sentence: CodeStable is an AI coding workflow for serious engineering, orchestrating the software lifecycle instead of Agents
- Quick overview of 6 entities + 3 processes (see "Quick system overview" above)
- Ask the user: "Where do you want to start most?" Provide three specific prompts:
- "I have a new feature to develop" → cs-feat
- "There's a bug in the code" → cs-issue
- "The project hasn't integrated CodeStable yet" → cs-onboard
Stop here, don't explain details of all sub-skills. Expand on specific ones only when the user asks.
退出
Exit
本技能没有"落盘"步骤。退出条件就一条:
- 已经告诉用户下一步触发哪个具体的 子技能(或确认用户只是来了解体系,没要做事)
cs-*
输出形如:
你这个诉求建议走——{一句话理由:因为它专门处理 X 类场景}。 触发后它会 {简述会发生什么:会先扫已有 spec / 会让你先描述 / 会进入分诊 / ...}。 现在切到cs-xxx吗?cs-xxx
确认后由用户触发那个子技能(或在同一会话里直接用对应技能名),本技能这次的工作结束。
This skill has no "documentation" step. The only exit condition is:
- Have told the user which specific sub-skill to trigger next (or confirmed that the user only came to learn about the system and doesn't want to perform tasks)
cs-*
Output example:
For this request, it is recommended to use— {one-sentence reason: because it specializes in handling X-type scenarios}. After triggering, it will {briefly describe what will happen: first scan existing specs / ask you to describe first / enter triage / ...}. Switch tocs-xxxnow?cs-xxx
After confirmation, the user triggers that sub-skill (or directly uses the corresponding skill name in the same session), and this skill's work for this round ends.
不做的事
Things not to do
- 不读写 下的内容产物——不打开 design.md、不改 items.yaml、不创建 feature 目录。这些是子技能的事
codestable/ - 不替子技能做决策——比如不在本技能里做 brainstorm 分诊(那是 的事),不在本技能里判 cs-arch 走哪个模式
cs-brainstorm - 不一次推荐多个技能——每次只指一条路。用户有两个独立诉求就分两轮:先做完 A 再回来做 B
- 不重复体系总览的细节——(onboard 后存在)才是权威完整版,本技能只放精简速读
codestable/reference/system-overview.md - 不绕过 ——仓库没接入就先 onboard,不要试图直接路由到下游
cs-onboard
- Do not read/write content products under — do not open design.md, modify items.yaml, or create feature directories. These are tasks for sub-skills
codestable/ - Do not make decisions on behalf of sub-skills — e.g., do not perform brainstorm triage in this skill (that's 's job), do not judge which mode cs-arch should use in this skill
cs-brainstorm - Do not recommend multiple skills at once — only point to one path each time. If the user has two independent requests, split into two rounds: finish A first then return to B
- Do not repeat details of the system overview — (exists after onboarding) is the authoritative full version; this skill only provides a concise overview
codestable/reference/system-overview.md - Do not bypass — if the repository is not integrated, onboard first, do not attempt to route directly to downstream skills
cs-onboard