browser-use
A library that lets AI agents control a real web browser -- clicking, typing, navigating and reading pages -- to complete tasks that require the open web.
- Actively maintained
- Permissive license
Overview
browser-use gives an LLM agent a structured way to perceive and act on a live browser: it extracts the interactive elements of a page, lets the agent decide what to click or type, and handles navigation, so tasks that require actually using a website (rather than calling an API for it) become automatable. It's become the most widely adopted library specifically for this niche.
browser-use is a fit for teams building agents that need to interact with websites that don't offer a usable API -- filling forms, comparing prices, navigating multi-step flows -- as a building block rather than a full no-code automation platform.
- Categories
- AI Agents & Agent Tools
- Languages
- Python
- License
- MIT
Related Projects
Skyvern
AGPL-3.0Automates browser-based workflows using computer vision and LLMs to understand and interact with websites, even ones it hasn't seen before.
Actively maintained
- AI Agents & Agent Tools
Composio
MITPowers 1000+ toolkits, tool search, context management and authentication so AI agents can actually take action, not just plan it.
Actively maintained
- AI Agents & Agent Tools
MCP Servers
MITOfficial reference implementations of the Model Context Protocol -- the open standard now adopted across OpenAI, Google, Microsoft and Anthropic for connecting agents to tools and data.
Actively maintained
- AI Agents & Agent Tools
LangGraph
MITBuild resilient, stateful agents as graphs -- explicit nodes and edges for control flow, rather than an implicit agent loop, for reliable long-running or multi-step workflows.
Actively maintained
- AI Agents & Agent Tools
Spotted an error? Suggest an edit on GitHub.