AbhishekRGoudar/Frontline-Worker-Support
GitHub: AbhishekRGoudar/Frontline-Worker-Support
Stars: 0 | Forks: 0
# Aegis AI Emergency Navigator
## Features
- Citizen case intake with SOS flow and browser geolocation
- Multi-agent workflow: triage, guidance, booking, follow-up, oversight
- Server-side Gemini calls so the API key is not shipped to the browser
- Overpass/OpenStreetMap facility search from the backend
- JSON persistence for cases and audit events in `data/app-data.json`
- 3D hospital navigator with department routing
- Degraded mode fallback when AI or network services are unavailable
## Tech Stack
- React 19, TypeScript, Vite
- Tailwind CSS
- Express API server
- Gemini via `@google/genai`
- Three.js and Motion
- OpenStreetMap Overpass API
## Run Locally
1. Install dependencies:
npm install
2. Create `.env` from `.env.example` and add `GEMINI_API_KEY` if you want AI responses. Without a key, the app uses local safety rules.
3. Start the app:
npm run dev
4. Open:
http://localhost:3000
## Scripts
- `npm run dev`: runs the Express server with Vite middleware
- `npm run build`: builds the frontend
- `npm run start`: serves the built frontend and API in production mode
- `npm run lint`: runs TypeScript checks
## Safety Notice
This is a prototype assistant for routing and workflow simulation. It does not replace emergency services, professional medical advice, or official dispatch systems. In a real emergency, call the local emergency number immediately.
## Next Production Steps
- Replace JSON storage with PostgreSQL, Supabase, Firebase, or MongoDB
- Add authentication and role-based access for citizens, dispatchers, hospitals, and admins
- Add real-time case updates with WebSockets or a realtime database
- Add automated tests for triage, booking, and navigator matching
- Add secure deployment configuration for frontend, backend, and secrets
标签:自动化攻击