A hybrid architecture platform for identifying misinformation and bias in web content.

• Extension: Chrome Extension (Manifest V3, React + Vite) - apps/extension
• Web: Next.js Dashboard & API Gateway - apps/web
• ML Core: Python FastAPI Service - services/ml-core

• Node.js 18+
• Python 3.10+
• Docker (optional, for ML service)

cd services/ml-core
pip install -r requirements.txt
uvicorn main:app --reload
# Service runs on http://localhost:8000

cd apps/web
cp .env.example .env.local
# Fill in Auth0 credentials in .env.local
npm install
npm run dev
# Web runs on http://localhost:3000

1. Build the extension:

   cd apps/extension
   npm install
   npm run build

2. Open Chrome -> Extensions (chrome://extensions/)
3. Enable "Developer mode"
4. Click "Load unpacked" and select apps/extension/dist

• The Extension communicates with the Web Platform (localhost:3000) for authentication and analysis proxying.
• The Web Platform proxies requests to the ML Service (localhost:8000).

See apps/web/.env.example for required keys.