GitDocs: Automated GitHub Repository Documentation

This repository contains the code for GitDocs, a tool that automatically generates documentation for GitHub repositories. GitDocs leverages large language models (LLMs) to analyze code, generate README files, add docstrings, and provide a conversational interface for interacting with the codebase.

Project Structure

├── gitdocs
│   ├── code_analyzer.py
│   ├── docstring_generator.py
│   ├── repository_scraper.py
│   ├── readme_generator.py
│   ├── chat_interface.py
│   ├── main.py
│   ├── code_extensions.txt
│   └── environment.yml
└── README.md

Dependencies

• Python 3.12+
• google-generativeai
• rich
• dotenv
• gitpython
• argparse

Installation

1. Clone the repository.
2. Install the required packages:

conda env create -f environment.yml
conda activate gitdocs

3. Create a .env file in the root directory and set the GEMINI_API_KEY environment variable. You can obtain a Google Generative AI API key from https://aistudio.google.com/app/apikey.

Usage

To generate documentation for a GitHub repository, run the following command:

python gitdocs/main.py <repository_url> [--generate-readme] [--generate-docstrings] [--chat] [--branch <branch_name>]

Example:

python gitdocs/main.py https://github.com/google/python-fire --generate-readme --generate-docstrings --chat --branch master

Flags:

• --generate-readme: Generates a README.md file for the repository.
• --generate-docstrings: Adds docstrings to the code files.
• --chat: Starts an interactive chat session where you can ask questions about the codebase.
• --branch <branch_name>: Specifies the branch of the repository to be used (default branch is used if unspecified).

Features

• Code Analysis: Uses a large language model to analyze the code and generate a structured representation of the codebase.
• README Generation: Automatically creates a comprehensive README.md file based on the code analysis.
• Docstring Generation: Adds docstrings to code files, explaining the functionality of functions, classes, and methods.
• Interactive Chat Interface: Provides a conversational way to explore the codebase, ask questions, and get insights.

Contributing

Contributions are welcome! Please feel free to open issues and submit pull requests.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for more information.

---

Generated with ❤️ using GitDocs.