An entry point for a bug-fixing sub-flow that moves from identifying an issue to verifying a fix, producing report, analysis, and fix-note artifacts. Triggers: user says "fix bug", "there's a problem", or "fix XX". Routes to report, analyze, or fix based on existing artifacts. Simple issues take the express lane.
This skill serves as an automated entry point for a bug-fixing sub-flow, guiding the AI agent from issue detection through to fix verification, while generating three key artifacts: a report, an analysis, and a fix-note. It is triggered by natural language phrases like "fix bug," "there's a problem," or "fix XX," making it intuitive for users to initiate. The skill acts as a routing engine, inspecting existing artifacts to determine whether to proceed with reporting, analysis, or direct fixing, enabling context-aware progression. For straightforward issues, it bypasses intermediate steps via an express lane, accelerating resolution. It is designed for integration with AI agent platforms such as Claude Code, Cursor, or Codex, and relies on project-specific file conventions to track the state of bug-related work. This approach streamlines debugging workflows by ensuring consistent documentation and structured problem-solving.
npx skills add https://github.com/liuzhengdongfortest/codestable/tree/main/skills/cs-issueInstalled? Explore more Research & Data Analysis skills: affaan-m/uspto-database, affaan-m/scholar-evaluation, affaan-m/literature-review · View all 6 →
Entry point for a bug-fixing sub-flow that moves from identifying an issue to verifying a fix, producing report, analysis, and fix-note artifacts. Triggered when the user says "fix bug", "there's a problem", or "fix XX". Acts as a router, selecting the appropriate step (report, analyze, fix) based on existing artifacts. Simple issues take the express lane.
This repo contains 20 individual skills — each has its own dedicated page.
CodeStable 工作流根入口，介绍体系全貌并把诉求路由到对应 cs-* 子技能。触发：用户只输入 `cs`、说"介绍一下 codestable"、"该用哪个技能"、"不知道用哪个"，或诉求还很开放未收敛。本技能只做路由不做事。
维护 `.codestable/architecture/` 这份只记现状的系统地图，三种模式 update / check / backfill。触发：用户说"刷新 architecture"、"做架构检查"、"补这个模块的架构文档"、"方案和代码对得上吗"，或 feature 阶段需要先做架构动作。不写未来规划（走 cs-roadmap）。
想法还模糊时的讨论入口，做分诊后路由到 feature-design / feature-brainstorm / roadmap。AI 是思考伙伴不是记录员。触发：用户说"有个想法还没想清楚"、"先 brainstorm 一下"、"聊一聊这块"、"方向还在摇摆"。不处理 bug 和重构。
把已拍板的技术选型、架构决定、长期约束、编码规约记成永久性决策文档（tech-stack / architecture / constraint / convention 四种）。触发：用户说"记录决定"、"归档技术选型"、"ADR"、"记录这条约束"、"把规约写下来"，或 design / analyze 后做出重要选择。只归档已拍板的，讨论中的不归档。
对仓库做定向代码探索并把"提问→读代码→得结论"沉淀为可检索证据，三种类型 question / module-overview / spike。触发：用户说"先 explore 一下"、"这个仓库里 X 怎么实现"、"快速熟悉这个模块"、"把探索结果存档"。
新功能开发的子流程入口，把"加个 X 能力"从想法走到验收闭环。触发：用户说"做新功能"、"加个 X"、"实现 XX"。只做路由，根据已有产物决定走 brainstorm / design / fastforward / implement / acceptance。不处理 bug。
feature 流程阶段 3——验收闭环：对照 design 核实现 + 回写 architecture / requirement / roadmap，最后产出 {slug}-acceptance.md。触发：用户说"功能写完了验收一下"、"做最后检查"、"准备 merge"、"出验收报告"。前置依赖 cs-feat-impl 完成。
feature 流程阶段 1——为新功能起草 {slug}-design.md 作为后续实现和验收的唯一输入，拍板后抽出 checklist。触发：用户说"开始设计方案"、"写 design doc"、"准备实现 XX"，前提是已知道做什么、为谁、怎么算成功。
feature 流程的超轻量通道——不写 design / checklist 直接动手，但先指引 AI 查 CodeStable 知识库再开工。触发：用户说"快速模式"、"fastforward"、"别那么多步骤"、"直接开干"，且需求小到不值得走 design 流程。
feature 流程阶段 2——按 {slug}-checklist.yaml 里 design 切好的 paradigm 维度 steps 推进，每步具体改哪个文件由 implement 自决，写完用统一格式汇报。触发：用户说"方案确认了开始实现"、"按方案写代码"、"开工"。前提是 design 已 approved 且有 checklist。遇到方案外情况要回方案谈不要硬冲。
写或更新对外指南文档——开发者指南（dev-guide）和用户指南（user-guide），产物在项目 docs/ 目录。任务导向（怎么用 X 做 Y），与 libdoc 的零件参考不同。触发：用户说"写文档"、"开发者指南"、"用户指南"，或 feature-acceptance 收尾时推送。
issue 流程阶段 2——读 report + 读代码定位根因、评估风险，给用户 2-3 个修复方案让 TA 拍板。这一步不改代码。触发：用户说"分析这个 bug"、"找根因"、"定位问题"，且已有 {slug}-report.md。
issue 流程阶段 3——按已确认根因和方案定点修复、验证、写 {slug}-fix-note.md 落档。两个入口：标准路径从 analyze 来，快速通道从 report 直接来。触发：用户说"开始修 bug"、"按分析修"、"动手改代码"。只动方案声明的文件，不顺手优化。
issue 流程阶段 1——通过对话把问题落成可复现、可追溯的 {slug}-report.md，并判定走标准路径还是快速通道。只问现象不猜根因。触发：用户说"提个 issue"、"记录这个 bug"、"我发现一个问题"。issue 工作流的起点。
把踩过的坑或好做法沉淀成可检索的 learning 文档，两条轨道 pitfall（坑）/ knowledge（默认做法）。触发：用户说"沉淀知识"、"learning"、"把这次经验记下来"，或 acceptance / fix 收尾时推送。
给库的公开表面（组件 / 函数 / 命令）逐条目生成参考文档，带清单追踪，支持单条目和批量。信息源是源码本身（与 guidedoc 任务导向不同）。触发：用户说"写 API 文档"、"组件文档"、"libdoc"，或 acceptance 后发现新增公开接口。
把新仓库或有零散文档的仓库接入 CodeStable 体系，两条路径自动判断：空仓库从零搭骨架，已有文档走审计 + 迁移映射。触发：用户说"在这个项目里用 CodeStable"、"搭 CodeStable 结构"、"初始化 CodeStable"、"迁移到 CodeStable"。
代码优化的子流程入口，处理"行为不变、结构变"的工作（结构 / 性能 / 可读性），按 scan → design → apply 分步执行每步人工放行。触发：用户说"优化一下 / 重构 / 重写 / 拆一下 / 性能不行 / 代码太长"且不夹带行为改动。不处理新需求 / bug / 跨模块架构重划。
refactor 流程的超轻量通道——直接识别 1-3 条低风险优化、一次确认、原地改、跑测试自证。触发：用户说"快速重构"、"小重构"、"简单优化下 XX 函数"、"别那么多步骤"，且改动在单函数 / 单组件局部、有测试可自证。
维护 `.codestable/requirements/` 下的能力愿景文档。三种模式 draft / backfill / update。触发：design 阶段起草愿景、acceptance 阶段落档，或用户说"刷新 requirements"、"补一份 req"、"先把愿景写下来"。
USPTO patent and trademark data workflow for official record lookup, PatentSearch queries, TSDR checks, assignment data, and reproducible IP research logs.
Structured scholarly-work evaluation for papers, proposals, literature reviews, methods sections, evidence quality, citation support, and research-writing feedback.
Systematic literature-review workflow for academic, biomedical, technical, and scientific topics, including search planning, source screening, synthesis, citation checks, and evidence logging.
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Agent skill repository discovered by 10x-chat research.
Review backend code for quality, security, maintainability, and best practices based on established checklist rules. Use when the user requests a review, analysis, or improvement of backend files (e.g., `.py`) under the `api/` directory. Do NOT use for frontend files (e.g., `.tsx`, `.ts`, `.js`). Supports pending-change review, code snippets review, and file-focused review.