Browser Use

Package

Open-source Python library and cloud for AI agents that use a browser

Price: Free, open source; cloud is paid
Access: LLM or Browser Use API key

About

Browser Use is an MIT-licensed Python library that lets an LLM drive a real browser to finish a task, plus a paid cloud with hosted agents and cloud browsers. It works with most LLM providers; cloud runs are billed per use.

What you can do with it

Let an LLM agent navigate sites, click and fill in forms from a prompt
Extract data from websites that need a login or JavaScript
Run hosted browser agents or stealth cloud browsers through an API

Get started

Install the library with pip install browser-use
Install Chromium with uvx browser-use install
Add BROWSER_USE_API_KEY or another LLM key to .env

Example

import asyncio
from browser_use import Agent, ChatBrowserUse
from dotenv import load_dotenv

load_dotenv()

async def main():
    agent = Agent(task="Find the number 1 post on Show HN", llm=ChatBrowserUse())
    await agent.run()

asyncio.run(main())

Details

Hosting: Hosted service, Runs locally
Available in: Worldwide
Official SDKs: Python, JavaScript/TypeScript
MCP server: Remote