This is a full-stack Retrieval Augmented Generation (RAG) App that helps browse and query new code-bases via chat. We generate and store embeddings for a given repository in advance, and then when a user asks a question, we retrieve relevant code snippets and provide them to the LLM as context.

1. Clone this repository:

   git clone https://github.com/iPrSri/RAG-Code-Repository-Assistant
   cd RAG-Code-Repository-Assistant
   

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`
   

3. Install the required packages:

   pip install -r requirements.txt
   

4. Run the script:

   streamlit run app.py
   

5. When prompted, enter your OpenAI API key and GitHub token.

After indexing, you can ask questions about a given repository. Type 'quit' to move to the next repository or exit the program.