codebase-mapping
代码库结构和依赖分析,以理解代码库的架构。当需要(1)生成文件树或结构图,(2)分析导入/依赖图,(3)识别入口点和模块边界,(4)理解不熟悉的代码库的整体布局,或(5)为更深入的建筑分析做准备时使用。
📖 中文导读:代码库结构和依赖分析,以理解代码库的架构。当需要(1)生成文件树或结构图,(2)分析导入/依赖图,(3)识别入口点和模块边界,(4)理解不熟悉的代码库的整体布局,或(5)为更深入的建筑分析做准备时使用。 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。
name: codebase-mapping description: Repository structure and dependency analysis for understanding a codebase's architecture. Use when needing to (1) generate a file tree or structure map, (2) analyze import/dependency graphs, (3) identify entry points and module boundaries, (4) understand the overall layout of an unfamiliar codebase, or (5) prepare for deeper architectural analysis.
Codebase Mapping
Maps repository structure and dependencies to enable targeted architectural analysis.
Quick Start
Generate a structural map:
python scripts/map_codebase.py /path/to/repo --output structure.json
Process
- Clone or access the target repository
- Generate file tree excluding noise (node_modules, pycache, .git, etc.)
- Parse imports to build dependency graph
- Identify entry points (main.py, index.ts, setup.py, pyproject.toml)
- Detect boundaries - package structure and public APIs
Output Artifacts
The skill produces:
file_tree.txt- Annotated directory structuredependencies.json- Import graph in adjacency list formatentry_points.md- Identified entry points with descriptionsmodule_map.md- Package boundaries and public interfaces
Key Patterns to Identify
Entry Point Detection
Look for these patterns:
- Python:
if __name__ == "__main__",setup.py,pyproject.toml - Node:
package.jsonmain/bin fields,index.js - Frameworks:
app.py(Flask),manage.py(Django),main.ts(Nest)
Dependency Classification
Classify imports as:
- External: Third-party packages (from package manager)
- Internal: Project modules (relative imports)
- Standard: Language standard library
Noise Exclusion
Always exclude:
node_modules/
__pycache__/
.git/
.venv/
venv/
dist/
build/
*.egg-info/
.mypy_cache/
.pytest_cache/
Integration with Other Skills
This skill provides the foundation for:
data-substrate-analysis→ Focus on types.py, models.pyexecution-engine-analysis→ Focus on runner filescontrol-loop-extraction→ Focus on agent.py, loop filescomponent-model-analysis→ Focus on base classes
Example Output
## Repository: langchain
### Structure Summary
- 342 Python modules across 28 packages
- Primary entry: langchain/__init__.py
- Core packages: agents, chains, llms, tools
### Key Files for Analysis
- Types: langchain/schema.py, langchain/types.py
- Execution: langchain/agents/executor.py
- Tools: langchain/tools/base.py