Browserbase MCP
Last updated December 15, 2024
About Browserbase MCP
An MCP server that enables AI assistants to control cloud browsers with Browserbase and Stagehand.
Browserbase MCP Server is an open-source Model Context Protocol implementation that provides AI assistants with browser automation through Browserbase and Stagehand. It allows LLMs to navigate web pages, extract structured data, capture screenshots, and perform automated interactions with atomic precision. Designed for AI-powered IDEs and agent workflows, the server supports multi-session management, parallel browser sessions, and annotated screenshot vision for handling complex DOMs. It works with popular AI models like Claude, OpenAI, and Gemini, making it a versatile tool for testing, data collection, and automated web interaction.
Key features
- Browser automation with Browserbase cloud instances
- Navigate, click, and fill forms programmatically
- Extract structured data from web pages
- Capture full-page and element-level screenshots
- Annotated screenshot vision for complex DOMs
- Manage sessions (create, reuse, close)
- Run multiple browser sessions in parallel
- Supports Claude, OpenAI, Gemini, and other LLMs
Related tools
An MCP server that provides browser automation through Puppeteer, enabling interaction with Chrome instances.
Enables AI models to browse, scrape, and interact with live websites in real time via MCP.
A Model Context Protocol (MCP) server for Ashra, enabling data extraction and web scraping.
An MCP server that enables natural language management of Neon Postgres databases via IDEs and AI assistants.
An MCP server that connects AI agents with Infobip's global communication APIs for omnichannel engagement.
