README.md

metadata

title: Chat with GitHub Repository
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: gradio
python_version: 3.11
sdk_version: 4.44.0
app_file: app.py
pinned: false

🤖 Chat with GitHub Repository

A powerful AI-powered application that allows you to analyze any GitHub repository and ask questions about the codebase in natural language!

🌟 Features

• Repository Analysis: Clone and process any public GitHub repository
• AI-Powered Chat: Ask questions about the code using natural language
• Smart Code Understanding: Uses advanced embeddings to understand code structure and context
• Source References: Get direct references to relevant code files
• Multiple File Types: Supports Python, JavaScript, TypeScript, Java, C++, Go, Rust, PHP, Ruby, Swift, Kotlin, Scala, Markdown, JSON, YAML, and more

🚀 How It Works

1. Enter Repository URL: Paste any public GitHub repository URL
2. Processing: The app clones the repo, extracts code files, and creates embeddings
3. Ask Questions: Chat with the AI about the codebase using natural language
4. Get Answers: Receive detailed answers with references to specific code files

💡 Example Questions

• "What is this project about?"
• "How is the code structured?"
• "What are the main functions/classes?"
• "How does authentication work?"
• "What dependencies does this project use?"
• "Are there any tests in this codebase?"
• "How is error handling implemented?"
• "What are the main API endpoints?"

🛠️ Technology Stack

• Frontend: Gradio for the user interface
• AI/ML: Groq API for fast LLM inference, Sentence Transformers for embeddings
• Vector Database: ChromaDB for storing code embeddings
• Code Processing: GitPython for repository cloning
• Language Models: Groq API with DeepSeek-R1-Distill-Llama-70B

📁 Supported File Types

The application processes the following file types:

• Programming Languages: .py, .js, .ts, .jsx, .tsx, .java, .cpp, .c, .cs, .go, .rs, .php, .rb, .swift, .kt, .scala
• Configuration: .json, .yaml, .yml, .toml
• Documentation: .md, .txt

🔧 Configuration

The app uses Groq's fast and reliable API for LLM inference and Sentence Transformers for embeddings. To get started:

1. Get a free API key from Groq
2. Set your API key in the environment variable GROQ_API_KEY
3. The app is pre-configured to use the powerful deepseek-r1-distill-llama-70b model

No additional setup required - just add your Groq API key!

📝 Usage Tips

• Repository Size: Works best with small to medium-sized repositories
• Processing Time: Larger repositories may take longer to process
• Question Quality: More specific questions tend to get better answers
• File Limits: Files larger than 1MB are skipped to ensure optimal performance

🤝 Contributing

This project is open source and contributions are welcome! Feel free to:

• Report bugs
• Suggest new features
• Submit pull requests
• Improve documentation

📄 License

This project is licensed under the MIT License.

---

Note: This application processes public repositories only. Private repositories require authentication tokens.