
Browser
The Mcp Browser Use server wraps the browser-use library as an MCP server, enabling AI assistants to control a web browser through tools for navigation, form filling, button clicking, and data extraction. It uses HTTP transport instead of stdio to reliably handle long-running browser automation tasks (30-120+ seconds) that would otherwise timeout, and includes a web UI for real-time task monitoring, a REST API, and support for skills and deep research workflows. This solves the problem of enabling AI models to perform complex web-based automation and information gathering tasks that require actual browser interaction.
- GitHub stars
- 933
- tools
- 8
- last updated
- Jan 27, 2026
Tools
8 toolsPublic tool metadata for what this MCP can expose to an agent.
run_browser_agentExecute browser automation tasks
Execute browser automation tasks
No parameter schema in public metadata yet.
run_deep_researchMulti-search research with synthesis
Multi-search research with synthesis
No parameter schema in public metadata yet.
skill_listList learned skills
List learned skills
No parameter schema in public metadata yet.
skill_getGet skill definition
Get skill definition
No parameter schema in public metadata yet.
skill_deleteDelete a skill
Delete a skill
No parameter schema in public metadata yet.
health_checkServer status and running tasks
Server status and running tasks
No parameter schema in public metadata yet.
task_listQuery task history
Query task history
No parameter schema in public metadata yet.
task_getGet full task details
Get full task details
No parameter schema in public metadata yet.
Sponsors
Readme
mcp-server-browser-use
MCP server that gives AI assistants the power to control a web browser.
Table of Contents
- What is this?
- Installation
- Web UI
- Web Dashboard
- Configuration
- CLI Reference
- MCP Tools
- Deep Research
- Observability
- Skills System
- REST API Reference
- Architecture
- License
What is this?
This wraps browser-use as an MCP server, letting Claude (or any MCP client) automate a real browser—navigate pages, fill forms, click buttons, extract data, and more.
Why HTTP instead of stdio?
Browser automation tasks take 30-120+ seconds. The standard MCP stdio transport has timeout issues with long-running operations—connections drop mid-task. HTTP transport solves this by running as a persistent daemon that handles requests reliably regardless of duration.
Installation
Details
- Repository
- github.com/saik0s/mcp-browser-use
- Categories
- Web & Browser Automation, Automation & Workflows, Data & Analytics