
Cs
Route vague or open-ended dev requests to the right CodeStable `cs-*` sub-skill after scanning `.codestable/` state.
Overview
cs is a journey-wide agent skill that routes open CodeStable requests to the correct `cs-*` sub-skill after scanning `.codestable/`—usable whenever a solo builder is unsure which workflow skill to trigger.
Install
npx skills add https://github.com/liuzhengdongfortest/codestable --skill csWhat is this skill?
- Unified entry when users only say `cs`, ask which skill to use, or have unconverged intent
- Mandatory scan: Glob `.codestable/`, Read `attention.md`, optional `system-overview.md`, list active `features/` `issues
- Outputs only a recommended `cs-*` trigger—no spec writes, no substituting for sub-skills
- Teaches CodeStable 7-entity model (requirements, architecture, roadmap, features, issues, refactors, audits, compound)
- Falls back to `cs-onboard` when the skeleton or `attention.md` is missing
- 7 entities modeled under `.codestable/`
- 3 process lanes in the CodeStable system map
Adoption & trust: 925 installs on skills.sh; 902 GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You know you want to work in CodeStable but your ask is vague or you do not know which `cs-*` skill owns the next step.
Who is it for?
Repos already on or adopting CodeStable where the next action depends on whether you are onboarding, scoping a feature, fixing an issue, or exploring the system map.
Skip if: Skipping `cs-onboard` when `.codestable/` or `attention.md` is missing, or expecting this skill to write specs or execute feature/issue flows itself.
When should I use this skill?
User inputs only `cs`, asks to introduce CodeStable, does not know which skill to use, or their request is still open and unconverged.
What do I get? / Deliverables
You receive a single recommended sub-skill to invoke plus a brief why, without the router mutating specs or running child workflows for you.
- Recommended `cs-*` sub-skill name
- Short routing rationale
- Optional CodeStable system quick-read
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Clarify whether a vague idea should become a roadmap item or a feature spec before any design work.
Match "add permission check" to the feature or issue track after scanning active folders under `.codestable/`.
Point audit findings toward `audits/` or compound learnings without starting a fix flow from the wrong skill.
Route production bug reports to `issues/` analyze/fix skills while respecting in-flight work names.
How it compares
Use as a directory router inside CodeStable, not as a replacement for task-specific `cs-*` implementation skills.
Common Questions / FAQ
Who is cs for?
Solo and indie builders using CodeStable in Claude Code, Cursor, or similar agents who need a single entry when they cannot name the right `cs-*` skill.
When should I use cs?
At validate when scoping whether a bug or feature belongs in `issues/` vs `features/`; during build/pm when starting ambiguous work; in operate when you need audits or compound knowledge routing; or anytime you only type `cs` or ask how CodeStable fits together.
Is cs safe to install?
It is read-oriented routing guidance; review the Security Audits panel on this Prism page and treat `.codestable/` reads as normal repo access your agent already has.
SKILL.md
READMESKILL.md - Cs
# cs ## 启动必读 开始任何判断或动作前,先读取 `.codestable/attention.md`;缺失则视为骨架不完整,提示先补齐或运行 `cs-onboard`,不要回退到外部 AI 入口文件。 `cs` 是 CodeStable 工作流家族的统一入口。用户开口大概率不会指名某个 `cs-xxx`——可能只说"我想加个权限校验"、"这个地方有 bug"、"介绍下 codestable",甚至只发一个 `cs`。本技能负责接住开放式输入,弄清意图,路由到对的子技能。 **两件事,仅此两件**: 1. 用户带具体诉求 → 匹配场景路由表,告诉用户该触发哪个 `cs-*`,并简单说明为什么 2. 用户想了解体系 / 说不清想做什么 → 给精简体系速读 + 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 `.codestable/` 下内容产物 / 不替子技能跑流程。产出只有"建议触发哪个子技能"。 --- ## 收到调用先做的扫描 回应前每次都做(几个 tool 调用就够): 1. **看仓库有没有接入 CodeStable**——`Glob .codestable/` 看顶层目录 2. **存在**——必须先 `Read .codestable/attention.md`(如果缺失提示骨架不完整,先补齐或重跑 `cs-onboard`);再 `Read .codestable/reference/system-overview.md`(如果有);`Glob` 一下 `features/` `issues/` `roadmap/` 看进行中的工作(拿目录名就够,不逐份读) 3. **不存在**——后面提示用户先走 `cs-onboard` 4. **看用户原话**——开放式还是带具体诉求?带诉求匹配路由表,没诉求给体系介绍 扫完才回应。让用户感觉你心里有数。 --- ## 体系一图速读(用户没具体诉求 / 让你介绍时讲这个) CodeStable 把开发活动建模成 **7 个实体 + 3 个流程**,所有产物聚在 `.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("接下来怎么做这块大需求 + 模块切 + 接口定") ├── features/ 新增能力 spec 聚合根(design / impl / accept) ├── issues/ 修 bug spec 聚合根(report / analyze / fix) ├── refactors/ 重构 spec 聚合根(beta) ├── audits/ 审计实体(主动扫描发现清单,不定修) └── compound/ 知识沉淀(learning / trick / decision / explore) ``` **三条流程**: - **新增能力**:`cs-feat-design` → `cs-feat-impl` → `cs-feat-accept`(想法模糊先 `cs-brainstorm` 分诊) - **修 bug**:`cs-issue-report` → `cs-issue-analyze` → `cs-issue-fix` - **重构**(beta):`cs-refactor` / `cs-refactor-ff` **横切**:流程跑完发现"值得记下来" → `cs-learn` / `cs-trick` / `cs-decide` / `cs-explore` 沉淀到 `compound/`。 **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),不是 Agent。人在环——程序员对整体把控负责,AI 是高效执行体。 > 项目已 onboard 的话更详细总览看 `.codestable/reference/system-overview.md`。 --- ## 场景路由表 匹配用户的话到表里某行,告诉用户:"你这个诉求建议走 `cs-xxx`,因为 {一句话理由}"。 | 用户说什么 / 想做什么 | 路由到 | |---|---| | 仓库还没有 `.codestable/` | **先 `cs-onboard`**——所有其他 cs-* 都依赖这个目录 | | 想法还模糊 / "有想法没想清楚" / "先聊聊" / "不知道是不是新功能" | `cs-brainstorm`(分诊后路由到 design / feature-brainstorm 落盘 / roadmap) | | 新功能 / "加个 X" / "实现 XX" | `cs-feat`(路由 design / ff / impl / accept) | | BUG / 异常 / 报错 / "这里不对" / "文档错了" | `cs-issue`(路由 report / analyze / fix) | | 代码优化 / 重构 / 重写(行为不变) | `cs-refactor` / `cs-refactor-ff` | | 摸代码 / "X 是怎么实现的" / 提问调研 | `cs-explore` | | 审查系统 / 扫描 bug / 审计代码 / "有哪些问题" / "哪里可以优化" | `cs-audit`(主动扫描发现,只列清单不定修) | | 补 / 更新需求文档 | `cs-req` | | 补 / 更新 / 检查架构文档 / "刷新架构 doc" / "做架构体检" | `cs-arch` | | 大需求拆解 / "我想要一个 X 系统" / 排期规划 / 模块拆分 + 接口契约 | `cs-roadmap` | | 技术选型 / 长期约束 / 编码规约 | `cs-decide` | | 踩坑回顾 / 经验总结 / "值得记下来" | `cs-learn` | | 可复用编程模式 / 库用法 / "以后做 X 就该这样" | `cs-trick` | | 一两行的项目注意事项 / 编译特殊设置 / 命令陷阱 / "记到 attention.md" | `cs-note` | | 开发者指南 / 用户指南 | `cs-guide` | | 库 API 参考 | `cs-libdoc` | | 用户在 feature / issue 流程中间问"下一步" | 路由到对应入口(`cs-feat` / `cs-issue`),让该入口判断当前阶段 | **判不出来 / 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。 --- ## 几种需要特别留心的情况 ### 仓库还没接入 任何 cs-* 流程但 `.codestable/` 不存在 → 说明这一点建议**先 `cs-onboard`**。不要直接路由到 cs-feat / cs-issue——它们的 SKILL.md 都假设 `.codestable/` 已存在。 ### 大需求被误当成 feature "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 → 不路由到 `cs-feat`,路由到 `cs-brainstorm`(大概率判 case 3 → `cs-roadmap`)或直接 `cs-roadmap`。理由:直接起 feature 会变成巨型 design 塞不下。 ### "改一下 X" 但 X 是已有功能 先问这是 **bug 修复**(X 现在表现错了)还是 **需求变更**(X 现在表现没错,但策略变了): - bug → `cs-issue` - 需求变更 → `cs-req` 改需求 doc + 之后 `cs-feat` 跑实现 ### 进行中的工作 扫描看到 `features/` 或 `issues/` 下已有相关目录 → 提一句"看到 `features/2026-04-22-xxx/` 已经存在,是接着做这个吗?" 让用户确认续作还是开新的。 ### 沉淀类技能的细分 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` - 一两行常驻提示"CodeStable 技能每次启动都得知道 X" → `cs-note`(写到 `.codestable/attention.md`) 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 / 常驻提示} 哪一种?" --- ## 介绍模式(用户只说想了解 / 不知道做什么)