Skip to main content
MCP server CX Codex

ExecuteAutomation Playwright — Claude Code MCP server

ExecuteAutomation Playwright — an MCP server that connects Playwright to Claude, Cursor, GitHub Copilot, and other AI code editors.

no license updated 2026-10-29

Open on GitHub

What ExecuteAutomation Playwright does

The server receives commands from your agent and executes them in a real browser via Playwright.

Version 1.0.10 adds 143 device profiles—iPhone, iPad, Pixel, Galaxy, and desktop browsers with viewport emulation, user-agent headers, touch events, and device pixel ratios. Tell your agent "test on iPhone 13" or "rotate to landscape," and the server adapts.

Browsers install automatically on first run. Logs go to ~/playwright-mcp-server.log to keep JSON-RPC clean in stdio mode. HTTP mode lets you run it as a standalone server for headless systems or IDE worker processes.

What it helps with

  • Automate UI tests using natural language
  • Screenshot a website and check how it looks on different devices
  • Extract data from a page without writing code
  • Execute JavaScript in a real browser context
  • Generate test code from scenario descriptions

Example prompts for the agent

  • Open example.com and check if the login button is visible
  • Test the form on iPhone 13, click the button, and show me a screenshot
  • Scrape all links from the page and return the list

How to install ExecuteAutomation Playwright

Install

The MCP registry has no launch command for it: see the repository README

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about ExecuteAutomation Playwright

How do I install ExecuteAutomation Playwright in Claude Code?

Run claude mcp add --transport stdio playwright npx @executeautomation/playwright-mcp-server. Or via npm: npm install -g @executeautomation/playwright-mcp-server. Browsers install automatically on first use.

What does ExecuteAutomation Playwright do?

It connects Playwright to AI code editors via MCP, so your agent can control a browser: open websites, take screenshots, scrape pages, run tests, and execute JavaScript.

Does it support testing on mobile devices?

Yes. Version 1.0.10 includes 143 device profiles: iPhone, iPad, Pixel, Galaxy, and more. You can emulate any of them with correct viewport, user-agent, and touch events.

Similar MCP servers

All MCP servers for Browser and scraping →

Project sponsors

Companies that keep this analytics open