This project is a local AI-powered personal assistant that:

• Uses LLaMA 3 (via Ollama) as the local LLM
• Integrates with LangChain agents to summarize Gmail and Google Calendar
• Has a desktop UI built with Electron
• Uses FastAPI as the backend API server
• Returns responses in Markdown format for rich UI rendering

    ┌────────────┐         ┌────────────┐
    │  Electron  │  ─────▶ │  FastAPI   │
    │  Frontend  │         │  Backend   │
    └────────────┘         └────┬───────┘
                                 │
                 ┌────────────┐  │
                 │ LangChain  │◀─┘
                 │ Agents     │
                 └────┬───────┘
                      │
               ┌────────────┐
               │ LLaMA 3 LLM│
               │ via Ollama │
               └────────────┘

• 🔌 Local-only LLM via Ollama
• 📅 Summarize today's Google Calendar events
• 📬 Summarize unread Gmail emails
• 🧠 LangChain agents with tool calling
• 💬 Frontend sends prompt → backend handles agent logic → returns Markdown summary

cd frontend
npm install
npm start

ollama run llama3

cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

uvicorn app.main:app --reload