Agentic RAG PDF Chatbot

A powerful chatbot application that uses Agentic RAG (Retrieval Augmented Generation) to intelligently answer questions about uploaded PDF documents. The system combines the power of LangChain, OpenAI's GPT models, and vector storage to provide accurate, context-aware responses.

Features • Technology Stack • API Documentation • Setup • Usage • Security

Features

• 📄 PDF Document Upload & Processing
• 💬 Interactive Chat Interface
• 🔍 Intelligent Document Search using RAG
• 🧠 Context-Aware Responses
• 📊 Conversation History Management
• 🎭 Customizable Assistant Personalities
• 🔒 Secure Document Handling
• 🚀 RESTful API Architecture

Technology Stack

Backend

• FastAPI - Modern web framework for building APIs
• LangChain - Framework for developing applications powered by language models
• OpenAI GPT-4o - Advanced language model for generating responses
• FAISS - Vector storage for efficient document retrieval
• PyPDF - PDF document processing

Frontend

• Streamlit - Interactive web interface
• Python Requests - HTTP client for API communication

API Documentation

The backend provides a RESTful API with the following endpoints:

Conversation Management

Create New Conversation

POST /conversations/new

Creates a new conversation session and returns a unique conversation ID.

Response:

{
    "conversation_id": "uuid-string",
    "message": "New conversation created successfully"
}

Upload Documents

POST /conversations/{conversation_id}/upload

Upload PDF documents for processing in a specific conversation.

Parameters:

• conversation_id (path): UUID of the conversation
• files (form-data): List of PDF files to upload

Response:

{
    "message": "Documents uploaded and processed successfully."
}

Chat with Documents

POST /conversations/{conversation_id}/chat

Send a question and receive an AI-generated response based on the uploaded documents.

Request Body:

{
    "question": "Your question here",
    "assistant_name": "AI Assistant",
    "assistant_behavior": "Professional",
    "custom_instructions": ""
}

Response:

{
    "answer": "AI-generated response based on document context"
}

Additional Endpoints

• GET /conversations - List all conversations
• GET /conversations/{conversation_id}/history - Get chat history
• GET /conversations/{conversation_id}/files - List uploaded files
• GET /health - API health check
• POST /conversations/{conversation_id}/load - Load conversation documents

Setup & Installation

1. Clone the repository:

git clone https://github.com/mostafa-ghaith/agentic-rag-app.git
cd agentic-rag-app

2. Set up environment variables: Create a .env file in the root directory with the following variables:

OPENAI_API_KEY=your_openai_api_key

3. Using Docker (Recommended):

docker-compose up --build

4. Manual Setup:

Backend:

cd backend
pip install -r requirements.txt
uvicorn api.main:app --reload

Frontend:

cd frontend
pip install -r requirements.txt
streamlit run app.py

Usage

1. Access the web interface at http://localhost:8501
2. Upload PDF documents using the sidebar and click on "Process Documents"
3. Set the custom behaviour of the chatbot as wanted (Optional)
4. Start chatting with the bot about your documents
5. You can navigate to previous conversations and continue them from the sidebar

Chatbot Interface

Interactive chat interface with document upload and customizable assistant behavior

Development

To contribute to the project:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

Security

• API keys and sensitive data are managed through environment variables
• Document storage is temporary and session-based
• Secure file handling and validation
• Rate limiting on API endpoints
• Input sanitization and validation
• CORS policy implementation
• Regular security updates and dependency scanning

License

MIT License

Copyright (c) 2025 Mostafa Ghaith

Contact

Mostafa Ghaith
linkedin.com/in/mostafa-ghaith
github.com/mostafa-ghaith