
browser-use
22 Tools · 1 Intel
Browser Use is an open-source web-agent framework founded in 2024 by Magnus Müller and Gregor Žunić, two ETH Zurich data science graduates who built the initial demo in five weeks. The Y Combinator W25 company converts web page content into a format LLMs can act on — rather than relying on vision — and quickly became foundational infrastructure for AI browser automation, reaching 50,000 GitHub stars within three months of launch and raising a $17 million seed round.
Tools
Python library that lets AI agents control web browsers, automating form filling, shopping, and data extraction.
Self-healing harness from Browser Use that gives LLMs a reliable interface for navigating websites autonomously.
Experiment from Browser Use in which a coding agent executes plain-English video edits across a timeline.
Browser-based control panel for the Browser Use library, running AI agents that navigate websites for you.
Creates and runs multi-step browser workflows powered by LLMs, with reusable and replayable agent actions.
MCP server that runs automated QA on web apps with Browser-Use agents that crawl the site and report bugs.
Agent framework that runs natively in the browser, distributed as a TypeScript package.
Self-healing browser harness that lets LLM agents complete web tasks reliably, written in TypeScript.
Desktop app for running Browser Use AI browser agents locally, with a native UI for managing tasks and tools.
Type-safe Python wrapper around the Chrome DevTools Protocol for driving Chromium browsers programmatically.
SDK with programmatic primitives for building LLM-driven browser agents that automate real Chromium sessions.
Drives the macOS Accessibility API so AI agents can operate native Mac applications.
Automates QA by handing test flows to browser agents that exercise web apps end to end.
Python event bus with async and sync handlers, FIFO and concurrency options, and WAL persistence.
Community nodes that let n8n workflows orchestrate browser automation through the Browser-Use agent framework.
TypeScript client for the Browser Use Cloud API, covering creation and orchestration of browser agents.
Ready-made templates for the browser-use CLI, covering scheduler and Telegram automation flows.
Sample projects showing what the Browser Use automation framework can do, in Python and TypeScript.
Reference Next.js chat app showing the Browser Use v3 SDK driving a browser-controlling AI agent.
Go daemon that bridges Chrome DevTools Protocol, a port of harnessless for concurrent browser automation.
Proposal for websites to publish callable tools that AI agents can invoke directly in the browser.
Studio for designing and running browser-use AI agents that carry out automated tasks on the web.
Intel
browser-use 0.13.10 migrates to MCP Python SDK 2.1.1, pins dependencies to resolve three pypdf CVEs, and changes unknown MCP tool calls to report as application errors instead of silently succeeding.