Code Repository Analysis Chatbot

A full-stack application that allows users to analyze and chat about code repositories using Google's Generative AI and LangChain. The system clones GitHub repositories, processes the code files, and enables intelligent conversations about the codebase.

Screenshots

Repository Analysis Interface

Chat Interface with AI

Features

• Repository cloning and code analysis
• Intelligent code-aware chatbot using Google's Generative AI
• Vector database for efficient code search and retrieval
• FastAPI backend with RESTful endpoints
• React frontend for user interaction
• Support for Python, HTML, and CSS file analysis
• Conversation memory for contextual responses

Application Workflow

graph TD
    A[User Input] --> B[Repository URL]
    B --> C[Clone Repository]
    C --> D[Process Code Files]
    D --> E[Generate Embeddings]
    E --> F[Store in Vector DB]
    F --> G[Enable Chat Interface]
    G --> H[User Questions]
    H --> I[AI Responses]

Loading

Architecture

Backend Components

1. Repository Handler (repo_handler.py)

   • Manages Git repository cloning
   • Handles local repository storage

2. Document Processor (document_processor.py)

   • Processes code files using LangChain's document loaders
   • Supports Python, HTML, and CSS parsing
   • Implements intelligent text splitting for optimal processing

3. Vector Database Manager (vector_database_manager.py)

   • Manages Chroma vector database
   • Utilizes Google's Generative AI embeddings
   • Persists processed documents for quick retrieval

4. Chatbot Handler (chatbot_handler.py)

   • Implements conversation chain using LangChain
   • Maintains conversation memory for context
   • Uses Google's Gemini 1.5 Pro model for responses

Frontend

• React-based user interface
• Located in my-react-app directory

Setup

Prerequisites

• Python 3.8+
• Node.js and npm
• Google Cloud API credentials
• Git

Environment Variables

Create a .env file with the following variables:

GOOGLE_API_KEY=your_google_api_key

Installation

1. Clone the repository:

git clone https://github.com/charans2702/Code-Repository-Analysis-Chatbot.git
cd Code-Repository-Analysis-Chatbot

2. Install backend dependencies:

pip install -r requirements.txt

3. Install frontend dependencies:

cd my-react-app
npm install

Running the Application

1. Start the backend server:

uvicorn main:app --reload

2. Start the frontend development server:

cd my-react-app
npm run dev

API Endpoints

POST /initialize

Initializes a new repository for analysis

{
    "repo_url": "https://github.com/username/repo"
}

POST /chat

Sends a question to the chatbot

{
    "question": "What does this repository do?"
}

GET /status

Checks if a repository has been initialized

{
    "initialized": true
}

Technical Details

• Uses MMR (Maximum Marginal Relevance) for document retrieval
• Chunk size of 500 characters with 20 character overlap for document splitting
• Supports multiple programming language parsing
• Implements CORS middleware for cross-origin requests
• Uses Pydantic for request/response validation

Dependencies

Backend

• FastAPI
• LangChain
• Google Generative AI
• ChromaDB
• GitPython
• python-dotenv

Frontend

• React
• Additional frontend dependencies (specified in package.json)

Future Improvements

• Support for more programming languages
• Advanced code analysis features
• User authentication
• Repository caching
• Real-time collaboration

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

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