Browser Use
Makes any website controllable by an AI agent in plain Python
View on GitHubWhat it solves
The web automation gap: teaching an LLM to operate a browser the way a human would, without writing custom Playwright scripts for every site and task.
Overview
Browser Use gives AI agents a real browser to control, clicking, filling forms, navigating pages, extracting content. Built on Playwright, it bridges the gap between 'the model knows what to do' and 'the model can actually do it on a live web page.' Returns structured results from page interactions with a full action history the agent can reflect on. Supports any OpenAI-compatible model.
Key facts
- Language
- Python
- License
- MIT
- Maturity
- Fast-moving
- Maintainer
- browser-use
- Reviewed
- 2026-06-11
Where to start
The simple example shows the agent-browser loop in 15 lines (natural language task to structured page result) before adding memory or multi-step flows.
examples/simple.pyTechnologies
More in this domain
LangGraph
Stateful multi-agent orchestration for LLM applications
The state management problem in multi-agent systems: how you build agents that pass context across turns without losing the execution thread.
AutoGen
Multi-agent conversation framework from Microsoft Research
How to coordinate multiple specialized agents toward a shared goal without writing low-level orchestration logic by hand.
CrewAI
Role-based multi-agent orchestration for collaborative task execution
The boilerplate problem in multi-agent setup: defining roles, delegation rules, and task routing without writing a custom orchestration graph.