Playwright

browserautomationplaywrighttestingofficial

Summary

Browser automation for agents — navigate, click, type, and read pages using Playwright's accessibility tree, no screenshots required.

Features

Drive a real browser
Navigate, click, and type
Read pages via accessibility tree
Fill and submit forms
Deterministic, screenshot-free automation

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

Open your agent's MCP configuration file.
Merge the snippet below into it.
Restart the agent — the "Playwright" tools will be available.

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "@playwright/mcp@latest"
      ]
    }
  }
}

Claude Code

Add the server with a single CLI command.

claude mcp add playwright -- npx @playwright/mcp@latest

Claude Desktop

Add the server to claude_desktop_config.json (Settings → Developer → Edit Config).

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "@playwright/mcp@latest"
      ]
    }
  }
}

Cursor

{
  "command": "npx",
  "args": [
    "@playwright/mcp@latest"
  ]
}

Description

Microsoft's official Playwright MCP server. It lets an agent drive a real browser through Playwright — navigating, clicking, filling forms, and reading page content via the accessibility tree — enabling reliable web automation and end-to-end testing without relying on pixel screenshots.