gj-codebase-map
Map an existing repository into confirmed engineering standards and focused AI context. Use when onboarding a GitLab project, refreshing development or test standards, deriving docs/context and docs/modules, generating .gj/context.yml drafts, or scanning changed paths after major refactors.
Works with
Agent Skills format with YAML frontmatter. Claude Code reads it as-is.
---
name: "gj-codebase-map"
description: "Map an existing repository into confirmed engineering standards and focused AI context. Use when onboarding a GitLab project, refreshing development or test standards, deriving docs/context and docs/modules, generating .gj/context.yml drafts, or scanning changed paths after major refactors."
license: "MIT"
---
# GJ 代码库建图
## 工作流程
1. 使用 `rg --files` 了解仓库结构。
2. 从四个视角扫描:
- 技术:技术栈、运行时、依赖、外部集成。
- 架构:入口、分层、数据流、模块边界。
- 质量:约定、测试、CI、错误处理。
- 风险:安全、脆弱区域、性能、未知项。
3. 把扫描结果直接路由到后续流程会读取的长期文档,不创建中间扫描目录:
- 技术栈、构建命令、架构边界、目录和前后端约定:
`docs/standards/01-development-standard.md`;
- 测试框架、目录、命令、覆盖和 CI 规则:`docs/standards/07-test-standard.md`;
- API、数据库、安全或环境专项事实:分别写入
`docs/standards/03-api-standard.md`、`docs/standards/04-database-standard.md`、
`docs/standards/08-security-standard.md`、`docs/standards/10-environment-standard.md`,
以及按需的 API、数据库或技术方案文档;
- 模块结构、集成和依赖:`docs/context/module-map.md`、`docs/modules/*.md` 以及按需的
API/技术方案;
- 项目级长期限制:`docs/context/current-state.md`;可执行技术债和缺陷输出为
GitLab Issue 草稿,不写入仓库待办清单。起草前读取
`docs/standards/09-ai-development-boundary.md` 和
`docs/standards/11-notification-standard.md`。
4. 创建或更新当前上下文草稿:
- `docs/context/current-state.md`
- `docs/context/module-map.md`
- `docs/context/glossary.md`
- `docs/modules/*.md`
- `.gj/context.yml`
从 `.gj/doc-templates/module.md` 创建模块文档,使用语义化模块名,包含通用元数据契约,
并明确区分已观察事实和推断。
5. 根据模块路径更新 `.gj/context.yml`,让开发、计划和审阅 Skill 能加载模块文档、
功能文档和有效 ADR;`01`/`07` 由代码相关 Skill 固定读取,不要把所有规范加入
`always_load`。
6. 标记所有需要人工确认的推断,并列出 Dev Lead、QA、安全或模块负责人分别需要确认
的标准草稿。扫描到的现有习惯不自动等于正确规范。
7. 暂存或分享前扫描生成内容中的秘密信息。
## 规则
- 不创建 `docs/codebase/` 或其他扫描报告目录;扫描只是生成长期规范和上下文草稿的过程。
- 未经确认,不要把观察到的行为转为当前状态规则。
- `01-development-standard.md` 和 `07-test-standard.md` 中新增的项目规则先标记为待确认,
分别由 Dev Lead 和 QA/Dev Lead 确认后生效。
- 已有长期文档原地更新,不创建 `v2`、`final` 或 `new` 副本。
- 外部集成写入负责该集成的模块或技术/API 文档;没有长期解释价值时不额外建文档。
- 风险有明确修复动作时创建 Issue 草稿;只有后续任务都会依赖的当前限制才写
`current-state.md`。
- 发现 Token、私钥、密码或连接字符串时停止,并要求人处理。
## 输出格式
```markdown
## 代码库建图摘要
扫描路径:
创建/更新的开发与测试规范:
创建/更新的上下文和模块文档:
已观察事实:
待确认推断:
需要人工确认的规范:
风险、缺陷和建议的 GitLab Issue:
秘密扫描:
```
## 参考资料
需要查看小型示例建图时,读取 `references/demo-run.md`。More Refactoring skills
vercel-react-best-practices
vercel-labs/agent-skills
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
analyze-project
lllllllama/rigorpilot-skills
Rigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrypoints, review configs and insertion points, or flag suspicious implementation patterns without modifying code or running heavy jobs. Do not use for active command execution, broad refactoring, speculative code adaptation, or automatic bug fixing.
request-refactor-plan
mattpocock/skills
Create a detailed refactor plan with tiny commits via user interview, then file it as a GitHub issue. Use when user wants to plan a refactor, create a refactoring RFC, or break a refactor into safe incremental steps.

