Puppeteer Browser Automation MCP server

The Puppeteer MCP Server is a Python implementation of the Model Context Protocol that enables language models to interact with web browsers. This server provides browser automation capabilities through Playwright, allowing LLMs to navigate web pages, capture screenshots, fill forms, and execute JavaScript in a real browser environment.

Prerequisites

• Python 3.8 or newer
• pip (Python package installer)

Installation

1. Install the required packages:

pip install -r requirements.txt

2. Install Playwright browsers:

playwright install

Starting the Server

Run the server directly with the following command:

python puppeteer_server.py

Claude Desktop Configuration

To configure the server with Claude Desktop, add the following to your Claude configuration file:

{
  "mcpServers": {
    "puppeteer": {
      "command": "python",
      "args": ["path/to/puppeteer.py"]
    }
  }
}

Available Tools

Browser Navigation

Navigate to any URL using the puppeteer_navigate tool:

{
  "name": "puppeteer_navigate",
  "arguments": {
    "url": "https://example.com",
    "timeout": 60000  // optional, defaults to 60000ms
  }
}

Taking Screenshots

Capture screenshots of the entire page or specific elements:

{
  "name": "puppeteer_screenshot",
  "arguments": {
    "name": "my_screenshot",
    "selector": "#specific-element",  // optional
    "width": 1280,  // optional, default: 1280
    "height": 720,  // optional, default: 720
    "timeout": 30000  // optional, defaults to 30000ms
  }
}

Clicking Elements

Click on page elements using CSS selectors:

{
  "name": "puppeteer_click",
  "arguments": {
    "selector": ".button-class",
    "timeout": 30000  // optional, defaults to 30000ms
  }
}

Filling Form Fields

Fill out input fields with text:

{
  "name": "puppeteer_fill",
  "arguments": {
    "selector": "#input-id",
    "value": "text to fill",
    "timeout": 30000  // optional, defaults to 30000ms
  }
}

Executing JavaScript

Run JavaScript code in the browser context:

{
  "name": "puppeteer_evaluate",
  "arguments": {
    "script": "document.title",
    "timeout": 30000  // optional, defaults to 30000ms
  }
}

Error Handling

The server provides detailed error messages for common scenarios:

• Navigation failures
• Element not found
• Timeout errors
• JavaScript execution errors
• Screenshot failures

Logging and Debugging

The server implements comprehensive logging with different levels:

• INFO: Standard operations
• ERROR: Operation failures
• DEBUG: Detailed execution information

Additional notes:

• Browser launches in non-headless mode for better debugging
• Default viewport size is 1280x720
• All timeouts are configurable
• Console logs are captured and stored
• Screenshots are stored in memory with base64 encoding

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.