Shopping Agent

Multi-platform shopping assistant that helps manage carts, track orders, and research products across Amazon, Swiggy, Blinkit, and Uber Eats.

Features

• Platform Connectors: Connect to multiple shopping platforms
• Cart Management: Add items, verify delivery addresses
• Order Tracking: Monitor order status and delivery
• Product Search: Search across platforms (integration in progress)

Tech Stack

• Backend: FastAPI + SQLAlchemy (async)
• Database: SQLite with aiosqlite
• Frontend: Jinja2 templates + vanilla JS
• Browser Automation: Integrates with UI-Agent for headless browser control

Setup

# Install dependencies
uv sync

# Run the server
uv run python main.py

Visit http://localhost:8080 to access the web interface.

Project Structure

shopping-agent/
├── app/
│   ├── api/            # FastAPI routes
│   ├── connectors/     # Platform connectors (Amazon, Swiggy, etc.)
│   ├── config.py       # Settings
│   ├── database.py     # SQLAlchemy setup
│   └── models.py       # Database models
├── templates/          # Jinja2 HTML templates
├── static/             # CSS and JS
└── main.py             # Application entry point

API Endpoints

• GET /api/connectors/ - List configured connectors
• POST /api/connectors/ - Add/update connector
• POST /api/products/search - Search products
• POST /api/carts/add - Add item to cart
• GET /api/carts/{platform} - Get cart contents
• POST /api/carts/{platform}/verify-address - Verify delivery address
• GET /api/orders/ - Get all orders

Integration with UI-Agent

The Shopping Agent uses the UI-Agent for browser automation since most platforms don't provide buyer-side APIs. The UI-Agent handles:

• Browser navigation and interaction
• Cart manipulation via headless browser
• Address verification on checkout pages

Status

• Backend API structure
• Web UI templates
• Database models
• UI-Agent integration for Amazon
• Swiggy connector implementation
• Blinkit connector implementation
• Uber Eats connector implementation