Skip to main content

Playwright MCP Server

The Playwright Model Context Protocol (MCP) server is a powerful solution for automating Browser and API testing using Playwright.

With the Playwright MCP server, you can:

  • Enable LLMs to interact with web pages in a real browser environment.
  • Perform tasks such as executing JavaScript, taking screenshots, and navigating web elements.
  • Seamlessly handle API testing to validate endpoints and ensure reliability.
  • Test across multiple browser engines including Chromium, Firefox, and WebKit.

Playwright MCP Server

Installation​

You can install Playwright MCP Server package using either npm, mcp-get, or Smithery:

Playwright MCP Tips

To get started more quickly on Playwright MCP Server, watch the videos mentioned in the footer of this page under Docs

Installing via NPM​

To install Playwright MCP for Claude Desktop automatically via Smithery:

npm install -g @executeautomation/playwright-mcp-server

Installing via Smithery​

To install Playwright MCP for Claude Desktop automatically via Smithery:

npx @smithery/cli install @executeautomation/playwright-mcp-server --client claude

You can type this command into Command Prompt, Powershell, Terminal, or any other integrated terminal of your code editor.

Installing via MCP-GET​

To install Playwright MCP for Claude Desktop automatically via Smithery:

npx @michaellatman/mcp-get@latest install @executeautomation/playwright-mcp-server

Configuring Playwright MCP in Claude Desktop​

Here's the Claude Desktop configuration to use the Playwright MCP server.

Modify your claude-desktop-config.json file as shown below

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["-y", "@executeautomation/playwright-mcp-server"]
    }
  }
}

What is Model Context Protocol​

This video should give you an high level overview of what Claude's MCP is and how helpful it will soon become for AI agents