Browser Use

// Overview

Browser Use is a Python library that connects a large language model to a real browser. You describe a task in plain language, and the agent navigates pages, clicks, types, and reads content to get it done.

It is built for developers who want to automate web workflows that are hard to script with fixed selectors, such as filling forms, gathering data across sites, or running through a multi-step flow. You bring your own model (Browser Use's own model, OpenAI, Anthropic, or Google), and the library handles the browser action loop.

Within the computer-and-browser-use space, it focuses on giving the model a browser action set with recovery loops. The 0.13 release adds a beta agent backed by a Rust core, while the original Python agent stays available under the same package.

// What it does

• Drives a real browser from natural-language tasks, no hand-written selectors required
• Model-agnostic: works with ChatBrowserUse, ChatOpenAI, ChatAnthropic, or Google models
• Beta agent (0.13) built on a Rust core with persistent tools and recovery loops
• BrowserProfile options like headless mode and allowed_domains to scope where the agent can go
• Async Python API that returns a run history you can inspect, including the final result
• Optional hosted cloud with stealth, proxy rotation, and 1000+ integrations for scaling

// Getting started

Install Browser Use with the native core runtime (Python 3.11+), set an API key, then run your first agent.

uv add "browser-use[core]"
# or: pip install "browser-use[core]"

# .env
BROWSER_USE_API_KEY=your-key
# GOOGLE_API_KEY=your-key
# ANTHROPIC_API_KEY=your-key

from browser_use.beta import Agent, BrowserProfile, ChatBrowserUse
import asyncio

async def main():
    agent = Agent(
        task="Find the number of stars of the browser-use repo",
        llm=ChatBrowserUse(model='openai/gpt-5.5'),
        browser_profile=BrowserProfile(
            headless=False,
            allowed_domains=["*.github.com"],
        ),
    )
    history = await agent.run()
    print(history.final_result())

if __name__ == "__main__":
    asyncio.run(main())

uv add "browser-use[core]"
browser

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Fill in a job application from a resume and personal details
• Add a shopping list of items to an online grocery cart
• Research and gather information across multiple websites for a task
• Automate a repeatable multi-step web workflow that is hard to script with fixed selectors

// How Browser Use compares

Browser Use alongside other open-source computer & browser use tools AI/TLDR tracks, ranked by GitHub stars.