Playwright Lighthouse MCP

🚀 Playwright-Lighthouse MCP Server

An MCP server that uses Playwright and Lighthouse to analyze website performance. Through the Model Context Protocol (MCP), LLMs can perform website performance analysis.

🚀 Quick Start

This server is designed to enable LLMs to perform website performance analysis via the Model Context Protocol (MCP). It combines Playwright and Lighthouse to provide comprehensive performance analysis and screenshot capture.

✨ Features

• Perform performance analysis using Lighthouse.
• Capture screenshots.

📦 Installation

Prerequisites

• Node.js 18 or higher.
• npm.

Installation Steps

# Clone the repository
git clone https://github.com/kbyk004/playwright-lighthouse-mcp.git
cd playwright-lighthouse-mcp

# Install dependencies
npm install
npx playwright install

# Build the project
npm run build

💻 Usage Examples

Debugging the MCP Server

npm run inspector

Integrating with an MCP Client

This server is designed for clients that support the Model Context Protocol (MCP). For example, it can be integrated with the desktop version of Claude.

Desktop Configuration Example

Add the following to the desktop configuration file (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "playwright-lighthouse": {
      "command": "node",
      "args": [
        "/path-to/playwright-lighthouse-mcp/build/index.js"
      ]
    }
  }
}

Available Tools

1. run-lighthouse

Run a Lighthouse performance analysis on the currently open page.

Parameter Description:

• url: The URL of the website to analyze.
• categories: An array of categories to analyze (default: ["performance"]).
  • Available categories: "performance", "accessibility", "best-practices", "seo", "pwa".
• maxItems: The maximum number of improvement items to display per category (default: 3, maximum: 5).

2. take-screenshot

Capture a screenshot of the currently open page.

Parameter Description:

• url: The URL of the website to capture.
• fullPage: If true, capture a screenshot of the entire page (default: false).

Output Format

The analysis results include:

• An overall score for each selected category, with a color indicator.
• Key improvement areas within each category.
• The path to the saved report file.

📄 License

This project is under the MIT License. For more details, see LICENSE.