Skip to main content

​
Playwright MCP

A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright through structured accessibility snapshots.

​
Why Playwright MCP?

Playwright MCP enables AI agents to interact with web pages through structured accessibility data instead of screenshots, making browser automation fast, deterministic, and LLM-friendly.

Fast & Lightweight

Uses Playwright’s accessibility tree, not pixel-based input. No vision models needed.

Deterministic

Avoids ambiguity common with screenshot-based approaches through structured data.

LLM-Friendly

Operates purely on structured data that LLMs can understand natively.

​
Key Features

Comprehensive Automation

Navigate, click, type, evaluate JavaScript, and more through a rich set of browser automation tools.

Multiple MCP Clients

Works with VS Code, Cursor, Claude Desktop, Windsurf, Goose, and any MCP-compatible client.

Flexible Configuration

Configure browser options, timeouts, network filtering, and more through a comprehensive config system.

Advanced Capabilities

Optional PDF generation, coordinate-based interactions, and test assertion tools.

​
Quick Example

Here’s how simple it is to get started with Playwright MCP: 
{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"]
    }
  }
}
Once configured, you can ask your AI agent to:
  • Navigate to websites and capture page snapshots
  • Click buttons, fill forms, and interact with elements
  • Take screenshots or generate PDFs
  • Verify page content and run tests
  • Execute custom JavaScript on pages

​
Get Started

Introduction

Learn about Playwright MCP and how it works

Quick Start

Get up and running in minutes

Installation Guide

Install for your MCP client

Configuration

Configure advanced options

​
Explore Documentation

Core Concepts

Understand MCP and browser automation fundamentals

Tools Reference

Browse all available automation tools

API Reference

Explore the programmatic API