web-scout-mcp

web-scout-mcp

MCP.Pizza Chef: pinkpixel-dev

Assistants often cannot browse, or return a link rather than what is on the page. This does both halves: it searches the web through DuckDuckGo and pulls the readable text out of a page, stripping menus and adverts. So you can ask a question needing current information and get an answer drawn from real pages. It works in Claude Desktop and Cursor, needs no account or key, and installs by pasting a short config entry.

Web/Research
Writing

Use This MCP server To

Search the web and summarise what I find Pull the readable text out of a page Check a fact that changed recently Compare what a few pages say Read an article without the clutter Find current information on a topic

README

Web Scout MCP Logo

Web Scout MCP Server

npm version License Node.js Version

An MCP server for web search using DuckDuckGo and content extraction, with support for multiple URLs and memory optimizations.

✨ Features

  • 🔍 DuckDuckGo Search: Fast and privacy-focused web search capability
  • 📄 Content Extraction: Clean, readable text extraction from web pages
  • 🚀 Parallel Processing: Support for extracting content from multiple URLs simultaneously
  • 💾 Memory Optimization: Smart memory management to prevent application crashes
  • ⏱️ Rate Limiting: Intelligent request throttling to avoid API blocks
  • 🛡️ Error Handling: Robust error handling for reliable operation

📦 Installation

Global Installation

npm install -g @pinkpixel/web-scout-mcp

Local Installation

npm install @pinkpixel/web-scout-mcp

🚀 Usage

Command Line

After installing globally, run:

web-scout-mcp

With MCP Clients

Add this to your MCP client's config.json (Claude Desktop, Cursor, etc.):

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

🧰 Tools

The server provides the following MCP tools:

🔍 DuckDuckGoWebSearch

Initiates a web search query using the DuckDuckGo search engine and returns a well-structured list of findings.

Input:

  • query (string): The search query string
  • maxResults (number, optional): Maximum number of results to return (default: 10)

Example:

{
  "query": "latest advancements in AI",
  "maxResults": 5
}

Output: A formatted list of search results with titles, URLs, and snippets.

📄 UrlContentExtractor

Fetches and extracts clean, readable content from web pages by removing unnecessary elements like scripts, styles, and navigation.

Input:

  • url: Either a single URL string or an array of URL strings

Example (single URL):

{
  "url": "https://example.com/article"
}

Example (multiple URLs):

{
  "url": [
    "https://example.com/article1",
    "https://example.com/article2"
  ]
}

Output: Extracted text content from the specified URL(s).

🛠️ Development

# Clone the repository
git clone https://github.com/pinkpixel-dev/web-scout-mcp.git
cd web-scout-mcp

# Install dependencies
npm install

# Build
npm run build

# Run
npm start

📚 Documentation

For more detailed information about the project, check out these resources:

  • OVERVIEW.md - Technical overview and architecture
  • CONTRIBUTING.md - Guidelines for contributors
  • CHANGELOG.md - Version history and changes

📋 Requirements

  • Node.js >= 18.0.0
  • npm or yarn

📄 License

This project is licensed under the MIT License.

Made with ❤️ by Pink Pixel
✨ Dream it, Pixel it ✨

web-scout-mcp FAQ

Can I use this to look things up on the web?
Yes — it searches through DuckDuckGo and can read the resulting pages for you.
Do I need an account or key?
No, it needs no account or paid search key, which is the main reason to pick it.
Can it read a page I give it?
Yes — you can point it at a link and it pulls out the readable text without menus and adverts.
Which apps does it work with?
It is documented for Claude Desktop and Cursor.
How hard is it to set up?
You paste a short config entry into your assistant's settings, so no coding is involved.