Browser-use is an open-source framework available in Python and TypeScript designed to enable AI agents to autonomously control web browsers. The library facilitates complex web automation tasks, such as booking appointments or navigating CAPTCHAs, by allowing users to connect large language models (LLMs) to browser instances. Developers can implement the library locally via a Python library (MIT-licensed) or utilize a fully hosted cloud solution. The cloud service offers managed infrastructure, including stealth browsers, residential proxies, and profile synchronization, at a cost of $0.02 per browser-hour. The framework supports various deployment patterns: a fully hosted API for end-to-end execution, a CLI (Browser Harness) for giving existing agents like Claude Code or Cursor browser access, and a Python library for custom integration. Users can leverage specialized models like the BU2 (bu-2-0) optimized for browser automation or connect external models via providers like OpenAI, Anthropic, or Google using the ChatBrowserUse gateway. Technical capabilities include custom tool registration through a Tools class, support for local hardware via Ollama, and the ability to reuse existing Chrome profiles to manage authentication sessions.

Read original