browser-use
ActiveDescription
browser-use enables browser automation for agents, allowing LLMs to understand pages and perform complex web interactions.
Key Features
- LLM-driven browser automation - let LLMs understand page content and perform clicks, typing, navigation
- Multi-model support - built-in ChatBrowserUse plus Gemini, Claude, GPT and other model backends
- CLI commands - open/state/click/type/screenshot commands for fast browser control from terminal
- Custom tool extension - register custom Python functions as callable tools for the agent
- Cloud stealth browsers - optional Browser Use Cloud for proxy rotation and captcha solving
- Template scaffolding - generate ready-to-run agent templates via uvx browser-use init
Use Cases
💡 Automate repetitive web tasks like job application forms and grocery ordering
💡 Build web scraping agents that navigate pages and extract target information
💡 Add browser operation capabilities to AI coding assistants like Claude Code
💡 Run end-to-end testing and UI verification for web applications at scale
💡 Create personal assistants for complex tasks like PC part comparison and flight search
Strengths & Limitations
✅ Strengths
- • Actively maintained, recent updates
- • High community interest (109.4k stars)
- • Permissive open-source license (MIT)
- • Responsive to issues, low backlog
Tags
Categories
Quick Start
# Install (requires Python >= 3.11)
uv init && uv add browser-use && uv sync
# Write your agent
# agent.py
from browser_use import Agent, Browser, ChatBrowserUse
import asyncio
async def main():
agent = Agent(
task="Find the number of stars of the browser-use repo",
llm=ChatBrowserUse(),
browser=Browser(),
)
await agent.run()
asyncio.run(main())