12 MCP servers for browser
Browser MCP servers give your AI the ability to see and interact with web pages. Using tools like Puppeteer, Playwright, or direct Chrome integration, these servers can navigate to URLs, click buttons, fill forms, take screenshots, and extract content from any website. This is essential for tasks that require interacting with web UIs or scraping data from pages that require JavaScript rendering.
The range of browser servers covers different needs: some focus on headless automation for testing and scraping, others provide cloud-hosted browsers for scale, and some let your AI control your actual Chrome session so it can see exactly what you see. There are also specialized servers for extracting YouTube transcripts, converting web pages to markdown, and running end-to-end browser tests.
Browser automation is one of the most powerful MCP capabilities. Use these servers to validate UI changes, monitor competitor websites, extract structured data from pages without APIs, automate repetitive web workflows, or give your AI the context it needs by letting it read the page you're looking at.
by Anthropic
Web content fetching and conversion for efficient LLM usage. Fetches URLs and converts HTML to clean markdown for AI consumption.
npx -y @modelcontextprotocol/server-fetchby Anthropic
Control a headless browser to navigate pages, take screenshots, fill forms, and extract content from any website.
npx -y @modelcontextprotocol/server-puppeteerby Microsoft
Official Playwright MCP server for browser automation. Navigate, click, fill forms, take screenshots, and run end-to-end tests.
npx -y @playwright/mcpby browsermcp
Local browser control that lets AI see and interact with your active Chrome tab. Click, type, scroll, and read page content in real time.
npx -y @nicepkg/browsermcpby WebDriverIO
WebDriverIO-powered browser and mobile automation. Control web browsers, iOS/Android apps, and handle touch gestures through AI.
npx -y @anthropic/webdriverio-mcpby co-browser
AI-driven browser automation using the browser-use library. Handles multi-step web tasks like form filling, navigation, and data extraction.
pip install browser-use-mcp-serverby Mendable
Web scraping, crawling, and search powered by Firecrawl. Extract clean content from any website with JavaScript rendering.
npx -y firecrawl-mcpby executeautomation
Playwright-based browser automation with screenshot capture, page navigation, and element interaction for AI-driven testing workflows.
npx -y @executeautomation/playwright-mcp-serverby Browserbase
Cloud browser automation powered by Browserbase. Run headless browsers at scale with session recording and debugging.
npx -y @browserbasehq/mcp-server-browserbaseby kimtaeyoon83
Extract transcripts and captions from YouTube videos. Get timestamped text from any public video for analysis or summarization.
npx -y mcp-server-youtube-transcriptby nichochar
Connect to Chrome DevTools for debugging, profiling, and inspecting web applications. Access console, network, and DOM.
npx -y chrome-devtools-mcpby urlbox
Generate and manage screenshots, PDFs, and videos of web pages. Includes AI-powered screenshot analysis and web content extraction as Markdown, metadata, and HTML via the Urlbox API.
npx -y @urlbox/urlbox-mcp-server