News Use

AI-Powered Personalized News Aggregation

Scrape, analyze, and curate custom news articles, and see what others are interested in!

Table of Contents

• Quick Start
• Project Structure
• Development
• Customization
• API Endpoints
• Environment Variables
• Contributing
• Tech Stack
• Roadmap
• License

Notes

• Hey all! This is Reagan. Welcome to the News Use repo! As of Oct 1. 2025, we aim to make this app more comprehensive and span across even more sources! One thing we loved about this project is that you get to see articles that other people made :D have fun generating!

Quick Start

# Install frontend dependencies
cd news-use
npm install

# Set up environment
cp .env.example .env.local
# Add your Convex URL (auto-generated on first run)

# Run development server
npm run dev

The app will open at http://localhost:5173

Backend Setup

cd backend

# Set up Python environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up environment
cp .env.example .env.local
# Add your GOOGLE_API_KEY and BROWSER_USE_API_KEY

# Run backend server
python api.py

The API will run at http://localhost:8000

Project Structure

news-use/
├── news-use/              # React frontend
│   ├── src/
│   │   ├── components/   # UI components
│   │   ├── lib/         # API client utilities
│   │   └── App.tsx      # Main application
│   └── convex/          # Convex backend
│       ├── schema.ts    # Database schema
│       └── newspapers.ts # CRUD operations
└── backend/             # FastAPI server
    ├── api.py           # Main API
    ├── news_scrapers/   # NYT & WashPost scrapers
    └── elaborators/     # AI summarization

Development

Frontend Development

cd news-use

npm run dev              # Start both frontend and Convex
npm run dev:frontend     # Start only frontend
npm run dev:backend      # Start only Convex
npm run build           # Production build
npm run lint            # Run linter

Backend API (Python)

cd backend

# Using uv (recommended)
uv pip install -r requirements.txt
python api.py

# Access API docs
# http://localhost:8000/docs

Database Management

Access Convex Dashboard to view/edit data:

cd news-use
npx convex dashboard

Customization

Add New News Sources

1. Create Scraper (backend/news_scrapers/source.py):

from browser_use_sdk import BrowserUse
from .models import Articles

client = BrowserUse(api_key=os.getenv("BROWSER_USE_API_KEY"))

def search_source(query: str) -> dict:
    task = client.tasks.create_task(
        task=f"Search for articles about: {query}",
        llm="gemini-flash-latest",
        schema=Articles,
    )
    result = task.complete()
    return result.output

2. Update API (backend/api.py):

from news_scrapers.source import search_source

@app.post("/search/source")
async def search_source_endpoint(search: SearchQuery):
    return search_source(search.query)

3. Update Frontend (news-use/src/lib/api.ts):

export async function searchSource(query: string) {
  const response = await fetch(`${API_BASE_URL}/search/source`, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ query }),
  });
  return response.json();
}

4. Add to QueryInput (news-use/src/components/QueryInput.tsx):

const sourceResponse = await searchSource(query);
allArticles.push(...sourceResponse.articles);

Modify AI Summarization Prompt

Edit the prompt in backend/elaborators/summarize_all.py (line ~44):

prompt = f"""
Based on these news articles, provide:

1. Your custom analysis requirements
2. Additional context you want
3. Specific focus areas
...
"""

Customize Newspaper Display

Modify sections in news-use/src/components/NewspaperDetail.tsx:

const parseSummary = (content: string) => {
  // Add your custom section parsing logic
  const customMatch = formatted.match(/5\.\s*Your Section[^]*/i);
  if (customMatch) {
    sections.push({
      title: "Your Custom Section",
      content: customMatch[0],
      key: "custom"
    });
  }
};

API Endpoints

FastAPI (Port 8000)

• POST /search/nyt - Search New York Times
• POST /search/washpost - Search Washington Post
• POST /summarize - Summarize articles with AI
• GET /health - Health check
• GET /docs - Interactive API documentation

Convex Functions

• createNewspaper - Save newspaper to database
• listNewspapers - Get public newspapers
• getNewspaper - Get specific newspaper by ID
• getStats - Get aggregated statistics

Environment Variables

Frontend (news-use/.env.local)

CONVEX_DEPLOYMENT=dev:your-deployment-name
VITE_CONVEX_URL=https://your-deployment.convex.cloud
VITE_API_URL=http://localhost:8000  # Optional, defaults to localhost:8000

Backend (backend/.env.local)

GOOGLE_API_KEY=your_google_api_key
BROWSER_USE_API_KEY=your_browser_use_api_key

CORS Issues

If deploying to production, update CORS in backend/api.py:

app.add_middleware(
    CORSMiddleware,
    allow_origins=["https://your-domain.com"],  # Replace *
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing)
3. Commit changes (git commit -m 'Add feature')
4. Push branch (git push origin feature/amazing)
5. Open Pull Request

Tech Stack

• Frontend: React, TypeScript, Vite, Tailwind CSS v4
• Backend: Convex, FastAPI, Render
• AI: Google Gemini Flash (summarization)
• Scraping: Browser Use SDK (automated browsing)

Roadmap

• Add more news sources (Reuters, Bloomberg, BBC)
• News analytics (sentiment analysis, trends, etc.)
• Email notifications for new articles
• Advanced filtering and search

License

MIT - See LICENSE file for details

---

Built with Browser Use - Automated web scraping powered by AI

Made with ❤️ by Reagan + Shawn + Browser Use