Documentation reference for writing Python code using the browser-use open-source library. Use this skill whenever the user needs help with Agent, Browser, or Tools configuration, is writing code that imports from browser_use, asks about @sandbox deployment, supported LLM models, Actor API, custom tools, lifecycle hooks, MCP server setup, or monitoring/observability with Laminar or OpenLIT. Also trigger for questions about browser-use installation, prompting strategies, or sensitive data handlin
Detected risks:
This skill provides comprehensive documentation and code reference for the browser-use open-source Python library, which enables developers to build AI agents that can interact with web browsers programmatically. It covers everything from initial setup and configuration to advanced features like custom tools, lifecycle hooks, and monitoring integration. The skill helps developers write Python code that imports from browser_use, configure Agent and Browser instances, and deploy to production environments including @sandbox.
The skill encompasses complete reference documentation for all major components: Agent configuration with prompting strategies and timeout handling, Browser setup with authentication and remote browser support, custom tool creation using the ActionResult system, the legacy Actor API for direct page manipulation, and integration with MCP servers and observability platforms like Laminar and OpenLIT. It provides detailed guidance on working with 15+ LLM providers, though it consistently recommends ChatBrowserUse as the default for optimal speed, cost, and accuracy.
This skill is designed for Python developers building browser automation solutions, AI agent developers working on web interaction tasks, data engineers scraping or testing web applications, and teams deploying browser automation to production. It explicitly focuses on the open-source library implementation and does not cover Cloud API/SDK usage, pricing information, or direct CLI-based browser automation, which are handled by separate specialized skills.
The browser-use library requires Python 3.11 or higher and is fully async. All entry points should use asyncio.run() to execute async functions.
Use uv for dependency management (not pip). Install with 'uv pip install browser-use' followed by 'uvx browser-use install' to set up browser dependencies.
ChatBrowserUse is the recommended default LLM provider, offering the fastest performance, lowest cost, and highest accuracy. The library supports 15+ LLM providers with detailed setup instructions available.
An API key is required for ChatBrowserUse and cloud features. Set the BROWSER_USE_API_KEY environment variable with a key obtained from https://cloud.browser-use.com/new-api-key.
Browser is simply an alias for BrowserSession - they are the same class and can be used interchangeably in your code.
Quick Setup:
.claude/skills/Repository
browser-use/browser-use