Skip to main content
Glama
faruklmu17

Playwright MCP Server

by faruklmu17

Playwright MCP Configuration

This repository contains everything needed to launch and connect to a working Playwright MCP (Model Context Protocol) server.

✅ Overview

This setup allows you to automate and inspect web pages using structured data instead of screenshots — ideal for LLM-powered tools like Claude, Cursor, and VS Code AI agents.


Related MCP server: MCP Playwright Server

🛠️ Prerequisites

  • Node.js v18 or newer

  • One of the following MCP-compatible clients:

    • VS Code or VS Code Insiders

    • Cursor

    • Claude Desktop


📁 Repository Contents

  • .vscode/settings.json — VS Code configuration to launch Playwright MCP

  • package.json — Includes required MCP dependencies

  • README.md — This file

  • .gitignore — Basic Node project ignore rules


🚀 Getting Started

1. Clone the Repository

git clone https://github.com/faruklmu17/playwright_mcp.git 
cd playwright-mcp-config

2. Install Dependencies

npm install

This installs all required packages, including @playwright/mcp.


3. Start the MCP Server

npx playwright-mcp --port 3550

You should see output like:

MCP Server started
Web server started

You can change the port if needed, e.g. --port=4000


4. VS Code MCP Client Configuration

Add this to your .vscode/settings.json:

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "@playwright/mcp@latest",
        "--port=3550"
      ]
    }
  }
}

This tells VS Code (or Cursor) how to launch and connect to the MCP server.


5. Optional: Use Persistent or Isolated Sessions

Persistent (default)

The browser state is preserved across sessions.

Isolated Example

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "@playwright/mcp@latest",
        "--isolated",
        "--storage-state=storage/state.json"
      ]
    }
  }
}

6. Additional Flags

You can enhance your MCP server using optional flags:

Flag

Description

--port <port>

Custom port (default is random/OS-assigned)

--headless

Run without browser UI

--device

Emulate a device (e.g., "iPhone 15")

--vision

Enable screenshot-based vision (if supported)

Check available flags with:

npx playwright-mcp --help

📅 Note About Browsers

You do not need to install playwright or run npx playwright install separately. The MCP package handles everything required for browser automation.


📄 .gitignore

node_modules/
.DS_Store

🤝 Contributing

Pull requests and suggestions are welcome!


📄 License

MIT

Related MCP Connectors

Related MCP Servers

  • -
    license
    Not graded
    quality
    Not graded
    maintenance
    Enables browser automation and web interaction through structured accessibility snapshots using Playwright. Provides fast, deterministic web page interaction without requiring screenshots or vision models.
    5,425,474 npm
    -
  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables AI assistants to control web browsers through Playwright automation, providing 50+ tools for navigation, interaction, testing, accessibility audits, and visual testing across Chromium, Firefox, and WebKit.
    6 npm
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    Provides browser automation capabilities for LLMs using Playwright, leveraging structured accessibility snapshots to interact with web pages without needing vision models. It enables tasks like web navigation, data extraction, and automated testing through a lightweight and deterministic toolset.
    3 npm
    Apache 2.0
  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables browser automation using Playwright, allowing LLMs to interact with web pages, take screenshots, generate test code, scrape web pages, and execute JavaScript in a real browser environment.
    11,800 npm
    MIT