AstroAI

AI-powered astrology demo combining a Flutter web frontend with a FastAPI backend that integrates Google's Gemini for astro features.

Monorepo Layout

backend/   # FastAPI app (Gemini-powered endpoints, SQLite, APScheduler)
frontend/  # Flutter web app (UI, calls FastAPI)

Quick Start

1) Backend (FastAPI)

cd backend
python -m venv .venv && source .venv/bin/activate  # optional but recommended
pip install -r requirements.txt
fastapi dev main.py
# API will run at http://localhost:8000

Endpoints exposed (high level):

• POST /horoscope
• POST /compatibility
• POST /natal_chart
• POST /review_event
• POST /with_celebrity
• GET /events-today
• POST /update-database/
• POST /admin/clear-past-events/

For details and request/response shapes, see backend/README.md.

2) Frontend (Flutter Web)

cd frontend/flutter
flutter pub get
flutter run -d chrome
# App will open in the browser (default port printed by Flutter)

The frontend calls the backend at http://localhost:8000 (see lib/services/api_service.dart). If you run on an Android emulator, change the base URL to http://10.0.2.2:8000.

How It Works

• Flutter UI collects basic user info (birth date, location) and stores it in app state.
• UI pages call FastAPI endpoints to fetch daily horoscope and today's events.
• The AI Assistant page uses the horoscope endpoint to provide a quick demo answer flow.

Troubleshooting

• CORS: CORS is enabled in backend/main.py for all origins. Lock it down as needed for production.
• Ports: Ensure the backend is running on 8000 and accessible from the browser/device you test with.
• Gemini API: The demo services use google-generativeai; ensure outbound internet access is available.

License

MIT. See LICENSE.