Google Drive Search AI Assistant

A production-ready full-stack AI application that provides a conversational interface to search files in a Google Drive folder using natural language queries.

Architecture

• Frontend: Streamlit web application
• Backend: FastAPI REST API
• AI Framework: LangChain with Gemini LLM
• Cloud Services: Google Drive API with Service Account authentication

Features

• Natural language file search in Google Drive
• Support for various search types:
  • Exact file name search
  • Partial file name search
  • MIME type filtering
  • Full-text content search
  • Modified date search
• Conversational AI interface powered by Gemini
• RESTful API for easy integration

Project Structure

project/
├── backend/
│   ├── main.py              # FastAPI application
│   ├── agent.py             # LangChain agent with Gemini
│   ├── drive_service.py     # Google Drive API service
│   ├── tools.py             # Search tools for the agent
│   ├── prompts.py           # System prompts
│   ├── requirements.txt     # Python dependencies
│   └── .env                 # Environment variables
├── frontend/
│   ├── app.py               # Streamlit application
│   └── requirements.txt     # Python dependencies
├── credentials/
│   └── service_account.json # Google Service Account key
└── README.md

Prerequisites

• Python 3.8+
• Google Cloud Project with Drive API enabled
• Service Account with Drive API access
• Gemini API key

Installation

1. Clone or Download the Project

cd /Users/nikhilcharantimath/Desktop/mini_ai

2. Set up Google Cloud Project

1. Go to Google Cloud Console
2. Create a new project or select an existing one
3. Enable the Google Drive API
4. Create a Service Account:
   • Go to IAM & Admin > Service Accounts
   • Create a new service account
   • Grant it the "Editor" role (or create a custom role with Drive read access)
   • Create a JSON key and download it
5. Share your Google Drive folder with the service account email

3. Configure Environment Variables

1. Copy the downloaded JSON key to credentials/service_account.json
2. Edit backend/.env:
   • Set GOOGLE_DRIVE_FOLDER_ID to your folder ID
   • Set GEMINI_API_KEY to your Gemini API key
   • Adjust HOST and PORT if needed

4. Install Backend Dependencies

cd backend
pip install -r requirements.txt

5. Install Frontend Dependencies

cd ../frontend
pip install -r requirements.txt

Running the Application

Start the Backend

cd backend
python main.py

The API will be available at http://localhost:8000

Start the Frontend

In a new terminal:

cd frontend
streamlit run app.py

The web interface will be available at http://localhost:8501

API Usage

Search Endpoint

POST /search
Content-Type: application/json

{
  "query": "find all PDF files modified today"
}

Response:

{
  "response": "Found 3 PDF files:\n- report.pdf (Modified: 2024-01-15)\n- document.pdf (Modified: 2024-01-15)\n..."
}

Search Query Examples

• "find file named 'budget.xlsx'"
• "find files containing 'meeting' in the name"
• "find all PDF files"
• "find Word documents modified today"
• "find files with 'important' in the content"
• "find recent spreadsheets"

Development

Backend Testing

cd backend
python -m pytest  # Add tests as needed

Frontend Development

The Streamlit app auto-reloads on code changes.

Security Notes

• Never commit the credentials/service_account.json file
• Keep your .env file secure
• Use environment-specific service accounts
• Regularly rotate service account keys

Troubleshooting

Common Issues

1. "Invalid credentials": Check your service account JSON and folder sharing
2. "API has not been used": Ensure Drive API is enabled in Google Cloud
3. "Folder not found": Verify the GOOGLE_DRIVE_FOLDER_ID in .env
4. "Gemini API error": Check your API key and quota

Logs

Check the terminal output for detailed error messages.

License

This project is for educational and demonstration purposes.