Understand your codebase with semantic search, context packs, and MCP integration. Free local embeddings included.
# Install from PyPI
pip install ai-code-autodoc
# Analyze your codebase
autodoc analyze ./src --save
# Semantic search with natural language
autodoc search "user authentication flow"
Search your code with natural language. Find functions by what they do, not just their names.
Group related code by feature. Perfect for giving AI assistants focused context.
9 tools for AI assistant integration via Model Context Protocol. Works with Claude, ChatGPT, and more.
ChromaDB with sentence-transformers. No API key required for semantic search.
Know which features are affected when files change. Perfect for CI/CD pipelines.
Dry-run mode, caching, and configurable models. Stay in control of LLM costs.
Group your code by feature for focused AI context
# autodoc.yaml
context_packs:
- name: auth
display_name: Authentication System
files:
- "src/auth/**/*.py"
- "api/routes/auth.py"
security_level: critical
tags: [security, core]
- name: payments
display_name: Payment Processing
files:
- "src/payments/**/*.py"
dependencies: [auth]
autodoc pack build auth --embeddings --summary
autodoc pack query auth "token validation"
autodoc pack diff auth
pip install ai-code-autodoc
autodoc analyze ./src --save
autodoc search "your query"
# Set API key for LLM features
export ANTHROPIC_API_KEY=sk-ant-...
# Generate AI summaries for packs
autodoc pack build --all --summary
9 tools for AI assistant integration
autodoc mcp-server