Browser Agents is a comprehensive web platform that lets you create, deploy, and manage AI-powered browser automation agents. Build intelligent bots that can navigate websites, extract data, fill forms, and perform complex web tasks using natural language instructions through a modern dashboard interface.
-
Clone and start:
git clone https://github.com/daviddaytw/browser-agents.git cd browser-agents docker-compose up --build
-
Access the platform:
- Dashboard: http://localhost:3000
- API: http://localhost:8000/docs
Browser Agents consists of two main components:
FastAPI microservice that handles browser automation:
- Technology: Python 3.11+, FastAPI, browser-use, Playwright
- Purpose: Core browser automation engine with REST API
- Features: Task execution, media capture, file handling
- Storage: In-memory task management with file system storage
Next.js web application for task management:
- Technology: Next.js 15, React 19, Material-UI, TypeScript
- Purpose: User interface for creating and monitoring tasks
- Features: Task templates, execution monitoring, user authentication
- Database: PostgreSQL with Drizzle ORM
- Open the dashboard at http://localhost:3000
- Create a task template describing what you want to automate
- Execute the task and monitor progress in real-time
- Download results and media captures
Configuration
OPENAI_API_KEY
- Your OpenAI API keyANTHROPIC_API_KEY
- Your Anthropic API keyDATABASE_URL
- PostgreSQL connection string
We welcome contributions! See our Contributing Guide for details.
Quick steps:
- Fork the repository
- Create a feature branch
- Make your changes
- Submit a pull request
Licensed under the Apache License 2.0.
- browser-use - The core browser automation library that powers our agents
Made with ❤️ by the David Day