Skip to content

MCP servers for browser automation

135 MCP servers that let AI control a web browser to navigate pages, fill in forms, take screenshots and scrape content. Each one shows how to install it in Claude, Cursor and VS Code, whether it runs locally or remotely, and the tools it provides.
  1. Headless Metabase MCP server using an API key, with 187 tools and read-only mode

    Runs locally
    • 9 tools
    • Open source
    • Browser automation
  2. Compare two logged-in identities to surface broken access control and IDOR

    Runs locally
    • Open source
    • Security
  3. One MCP proxy for Playwright, Figma, DevTools, Shopify Dev and Context7, with less output

    Runs locally
    • 2 tools
    • Open source
    • Developer tools
  4. Compose live parts of real web pages into one Mac workspace from your AI agent

    Not checked yet
    • Remote
    • Free
    • Productivity
  5. Let an AI assistant operate Java Swing and JavaFX desktop apps with no API needed

    Runs locally
    • 9 tools
    • Open source
    • Developer tools
  6. Find a verified web workflow and replay it with zero model calls, with failures shown

    Not checked yet
    • 2 tools
    • Remote
    • Free
    • Browser automation
  7. Run eight-view mobile checks on AI-generated web pages and get repair-ready evidence

    Runs locally
    • Open source
    • Developer tools
  8. Let an AI agent work visibly in your signed-in Chromium browser, locally, with audit

    Runs locally
    • Open source
    • Browser automation
  9. Let Claude Code, Cursor and Codex drive your own signed-in Chrome, with human fallback

    Runs locally
    • 1 tool
    • Open source
    • Browser automation
  10. Verify a running web app from inside: look, act, observe and assert, without screenshots

    Runs locally
    • 10 tools
    • Open source
    • Developer tools
  11. Rust browser automation over native webviews, for agents (MCP) and scripts

    Runs locally
    • Open source
    • Browser automation
  12. Puts browser and dev-server logs on one clock so agents can match errors to stack traces

    Runs locally
    • 11 tools
    • Free
    • Developer tools
  13. Gives an AI agent your real Chrome through compact text snapshots and a visible cursor

    Runs locally
    • Open source
    • Browser automation
  14. Chrome extension bridge that lets AI coding agents drive your already logged-in browser

    Runs locally
    • 9 tools
    • Free
    • Browser automation
  15. Let a coding agent read QA Wrap browser test sessions and failures, locally

    Runs locally
    • 4 tools
    • Open source
    • Developer tools
  16. Check what a website allows agents to do before retrying a blocked web step

    Not checked yet
    • 2 tools
    • Remote
    • Free
    • Developer tools
  17. Give coding agents their own background tabs in your signed-in Chrome

    Runs locally
    • Open source
    • Browser automation
  18. Shows which page text AI crawlers miss because it only appears after JavaScript runs

    Runs locally
    • 2 tools
    • Open source
    • Web search and scraping
  19. Capture website screenshots with local Chrome from a CLI or an AI agent

    Not checked yet
    • 1 tool
    • Remote
    • Open source
    • Browser automation
  20. Let an AI agent drive Android, iOS, TV and web apps, and save the steps as test suites

    Runs locally
    • 2 tools
    • Open source
    • Developer tools
  21. Hosted MCP server for running cloud browsers and managing Kernel apps from an agent

    Not checked yet
    • Remote
    • Open source
    • Browser automation
  22. Design guidance and headless Chromium QA for AI coding agents building front ends

    Runs locally
    • 13 tools
    • Open source
    • Browser automation
  23. MCP server and browser extension letting coding agents use your signed-in Chrome

    Runs locally
    • Open source
    • Browser automation
  24. Playwright MCP-compatible browser automation running on CloakBrowser Chromium

    Not checked yet
    • 2 tools
    • Remote
    • Open source
    • Browser automation
  25. Fail-closed browser engine for AI agents, MCP-native, without Chromium or CDP

    Runs locally
    • Open source
    • Browser automation
  26. Check pages for responsive design problems and screenshot them on real device sizes

    Runs locally
    • Open source
    • Browser automation
  27. Approve agent actions from your phone; sandboxed browser vault with saved logins

    Runs locally
    • Free
    • Security
  28. Local QA tools for coding agents: scan a page, inspect the UI, generate a Playwright repro

    Runs locally
    • 3 tools
    • Free
    • Developer tools
  29. Budgeted web page reading for AI assistants, read-only until you allow clicking

    Runs locally
    • 1 tool
    • Open source
    • Browser automation
  30. Give any agent your signed-in Mac browser, with a confirmation step before risky actions

    Runs locally
    • Free
    • Browser automation

See every kind of AI for browser automation: tools, agents, MCP servers and skills.Ranked: the best AI for browser automation.