AI-Doc-Gen: Multi-agent system that analyzes code, generates docs automatically

An AI-powered code documentation generator that automatically analyzes repositories and creates comprehensive documentation using advanced language models. The system employs a multi-agent architecture to perform specialized code analysis and generate structured documentation.

Read the full story behind this project:

• 🇺🇸 English: Docs That Don’t Rot: How Multi-Agent AI Rewrote Our Workflow
• 🇮🇷 از دستیار کدنویس تا همکار هوشمند؛ گام اول: کابوس مستندسازی

• Features
• Installation
• Quick Start
• Usage
• Configuration
• Architecture
• License

• Multi-Agent Analysis: Specialized AI agents for code structure, data flow, dependency, request flow, and API analysis
• Automated Documentation: Generates comprehensive README files with configurable sections
• GitLab Integration: Automated analysis for GitLab projects with merge request creation
• Concurrent Processing: Parallel execution of analysis agents for improved performance
• Flexible Configuration: YAML-based configuration with environment variable overrides
• Multiple LLM Support: Works with any OpenAI-compatible API (OpenAI, OpenRouter, local models, etc.)
• Observability: Built-in monitoring with OpenTelemetry tracing and Langfuse integration

• Python 3.13
• Git
• API access to an OpenAI-compatible LLM provider

1. Clone the repository:

2. Install using uv (recommended):

3. Or install with pip:

1. Set up your environment and configuration:

2. Run analysis and generate documentation:

Generated documentation will be saved to .ai/docs/ directory.

The tool automatically looks for configuration in .ai/config.yaml or .ai/config.yml in your repository.

• Exclude specific analyses: Skip code structure, data flow, dependencies, request flow, or API analysis
• Customize README sections: Control which sections appear in generated documentation
• Configure cronjob settings: Set working paths and commit recency filters

You can use CLI flags for quick configuration overrides. See config_example.yaml for all available options and .env.sample for environment variables.

The system uses a multi-agent architecture with specialized AI agents for different types of code analysis:

• CLI Layer: Entry point with command parsing
• Handler Layer: Command-specific business logic (analyze, document, cronjob)
• Agent Layer: AI-powered analysis and documentation generation
• Tool Layer: File system operations and utilities

• Python 3.13 with pydantic-ai for AI agent orchestration
• OpenAI-compatible APIs for LLM access (OpenAI, OpenRouter, etc.)
• GitPython & python-gitlab for repository operations
• OpenTelemetry & Langfuse for observability
• YAML + Pydantic for configuration management

This project is licensed under the MIT License - see the LICENSE file for details.

• Built with pydantic-ai for AI agent orchestration
• Supports multiple LLM providers through OpenAI-compatible APIs (including OpenRouter)
• Uses Langfuse for LLM observability