Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@pinkpixel/deep-research-mcp) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

🔍 Deep Research MCP

A Model Context Protocol (MCP) server for deep research on any topic. This tool enables AI assistants to search the web, extract content from webpages, and take screenshots.

✨ Features

🌐 Web Search: Search the web using Google
📄 Content Extraction: Extract and convert webpage content to Markdown
📸 Screenshot Capture: Take screenshots of webpages
💾 Content Saving: Save extracted content as markdown files
🖼️ Screenshot Saving: Save screenshots to user-accessible locations

📋 Requirements

Node.js 18 or higher
npm or yarn

🚀 Installation

# Install globally
npm install -g @pinkpixel/deep-research-mcp

# Or use with npx
npx @pinkpixel/deep-research-mcp

During installation, Playwright will automatically install Chromium, which is required for web automation.

🛠️ Usage

As an MCP Server

This package is designed to be used as an MCP server for AI assistants. You can configure it in your MCP configuration file:

{
  "mcpServers": {
    "deep-research": {
      "command": "npx",
      "args": [
        "-y",
        "@pinkpixel/deep-research-mcp"
      ]
    }
  }
}

Available Tools

The server provides the following tools:

1. Web Search

{
  "name": "web_search",
  "description": "Search the web",
  "arguments": {
    "query": "Your search query"
  }
}

2. Get Page Content

{
  "name": "get_page_content",
  "description": "Go to a webpage and extract its content",
  "arguments": {
    "url": "https://example.com",
    "takeScreenshot": true,
    "saveMarkdown": true,
    "saveScreenshot": true
  }
}

3. Take Screenshot

{
  "name": "take_screenshot",
  "description": "Take a screenshot of the current page",
  "arguments": {
    "saveScreenshot": true
  }
}

Saved Content

Extracted content and screenshots are saved to the following directories:

Markdown files: ./research-content/
Screenshots: ./research-content/screenshots/

These directories are created in the current working directory when the server starts.

🧩 Integration with AI Assistants

This MCP server is designed to work with AI assistants that support the Model Context Protocol. It provides tools for web research that can be used by AI models to gather information from the web.

🔒 Security and Privacy

The server runs in a sandboxed environment
Only HTTP and HTTPS protocols are supported
Bot protection detection is implemented to avoid being blocked
Content validation ensures only valid content is processed

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgements

Model Context Protocol for the protocol specification
Playwright for web automation
Turndown for HTML to Markdown conversion

📞 Contact

GitHub: pinkpixel-dev
Website: pinkpixel.dev

Made with ❤️ by Pink Pixel