Overview
Browserbase MCP Server is a cloud browser automation solution that connects LLM applications to live web pages via the Model Context Protocol. It enables LLMs to control cloud browsers, navigate pages, take screenshots, extract structured data, and perform automated actions with atomic precision, powered by Browserbase and Stagehand. The upgrade to Stagehand v3 delivers 20-40% faster core operations (act, extract, observe) through automatic caching, improved extraction across iframes and shadow roots, and streamlined schemas for more intuitive data extraction. New primitives — page, locator, frameLocator, and deepLocator — simplify automation across multiple engines, including Playwright, Puppeteer, and Patchright. The server supports both SHTTP (remote hosted URL) and STDIO transports, with optional experimental features, proxies, advanced stealth, and customizable viewport sizes (browserWidth and browserHeight). You can run locally via npm or Docker or connect remotely through Smithery’s hosted URL, providing appropriate API keys for Browserbase and any custom model provider. Features include multi-model support, vision-assisted screenshots, and context persistence for long-running sessions.
Features
Browser Automation
Control and orchestrate cloud browsers via Browserbase.
Data Extraction
Extract structured data from any webpage.
Web Interaction
Navigate, click, and fill forms with ease.
Screenshots
Capture full-page and element screenshots.
Model Flexibility
Supports multiple models (OpenAI, Claude, Gemini, and more).
Vision Support
Use annotated screenshots for complex DOMs.
Session Management
Create, manage, and close browser sessions.
High Performance
20-40% faster operations with automatic caching (v3).
