Rdvo MCP Server

🚀 Steel Golem Model Context Protocol Server

A model context protocol server that leverages the browser automation capabilities provided by Puppeteer and Steel. This server enables LLMs to interact with web pages, take screenshots, and execute JavaScript code, offering a real browser environment.

🚀 Quick Start

1. Install dependencies:

   npm install
   

2. Build the project:

   npm run build
   

3. Start the server:

   npm start
   

4. Open the Claude desktop application and start using it!

By default, the server will listen for requests on port 3000.

✨ Features

• Support browser automation operations using Puppeteer.
• Provide powerful console functions to interact with web pages.
• Support multiple scrolling behaviors and block positioning.
• Integrate the Steel protocol for model context interaction.
• Support custom user agent strings.
• Provide detailed logging and debugging options.

📦 Installation

Environment Variables

The following are the key environment variables required to run the server:

• NODE_ENV: Set to "production" to enable production mode. Default value: "development".
• PORT: Specify the port on which the server listens, ranging from 1024 to 65535. Default value: 3000.
• PUPPETEER_WS_ENDPOINT: Specify the Puppeteer WebSocket endpoint. Default value: ws://localhost:3000/ws/puppeteer.
• STEEL_API_KEY: The Steel API key for authentication.
• STEEL_SERVER_URL: The Steel service URL. Default value: http://localhost:8080.

Server Configuration

To customize the server behavior, edit the following configuration options:

{
  "puppeteer": {
    "defaultViewport": { // Default viewport settings
      "width": 1920,
      "height": 1080,
      "deviceScaleFactor": 1,
      "isLandscape": true,
      "hasTouch": false,
      "isMobile": false
    },
    "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.126 Safari/537.36" // Custom user agent string
  },
  "steel": {
    "apiEndpoint": "http://localhost:8080", // Steel API endpoint
    "timeout": 30000, // Request timeout in milliseconds
    "proxyEnabled": true, // Whether to enable proxy
    "proxyUrl": "http://localhost:8081" // Proxy server URL
  },
  "logging": {
    "enabled": true, // Enable logging
    "level": "info", // Log level: "debug", "info", "warn", "error"
    "outputFile": "./logs/server.log" // Log output file path
  }
}

📚 Documentation

Tools

• puppeteer_navigate

  • Navigate to any URL in the browser.
  • Input:
    • url (string, required): The URL to navigate to.
    • timeout (number, optional, default: 60000): The navigation timeout in milliseconds.
    • waitUntil (string, optional, default: "domcontentloaded"): The condition to consider the navigation successful. Options: "load", "domcontentloaded", "networkidle0", "networkidle2".

• puppeteer_screenshot

  • Take a screenshot of the entire page or a specific element.
  • Input:
    • name (string, required): The name of the screenshot.
    • selector (string, optional): The CSS selector of the specific element to screenshot.

• puppeteer_click

  • Click an element on the page.
  • Input: selector (string, required): The CSS selector of the element to click.

• puppeteer_fill

  • Fill a form field.
  • Input:
    • selector (string, required): The CSS selector of the form field.
    • value (string, required): The value to fill.

• puppeteer_select

  • Select an element in a SELECT tag.
  • Input:
    • selector (string, required): The CSS selector of the element to select.
    • value (string, required): The value to select.

• puppeteer_hover

  • Hover the mouse over an element on the page.
  • Input: selector (string, required): The CSS selector of the element to hover over.

• puppeteer_evaluate

  • Execute JavaScript code in the browser console.
  • Input: script (string, required): The JavaScript script to execute.

• puppeteer_get_content

  • Extract content from the current page.
  • Input: selector (string, optional): The CSS selector of the specific element to get content from. If not provided, the entire page content is returned.

• puppeteer_scroll

  • Scroll the page to trigger lazy loading.
  • Input:
    • behavior (string, optional, default: "auto"): The scroll behavior ("auto" or "smooth").
    • block (string, optional, default: "start"): The scroll block ("start", "end", "nearest", "center").

Resources

• Page
  • Provide access to the current web page.
• Screenshot
  • Return screenshot data.
• Title
  • Return the title of the current page.
• Content
  • Return the content of the current page.
• Cookies
  • Get or set the cookies of the current page.

🔧 Technical Details

Troubleshooting

• If you encounter Puppeteer issues, ensure that all necessary dependencies are installed on your system. Refer to the Puppeteer Troubleshooting Guide for more information.
• If you are using the Steel cloud service, ensure that the API key is valid and has the required permissions.
• If you are using a local Steel instance, ensure that it is running and accessible at the specified URL (if customized) or the default local address.

For more detailed configuration options and advanced usage, refer to the Steel documentation and Puppeteer API reference materials.