mcpField
W
Web MCP Server

Web MCP Server

Local
github.com/zgagnonSearch1Updated 77w ago
Repository

About Web MCP Server

Web MCP Server provides web search and page text retrieval for AI assistants. It combines DuckDuckGo result URLs with URL fetching and optional Mozilla Readability extraction.

Key features

  1. 01DuckDuckGo web search
  2. 02URL results with configurable limit
  3. 03Fetch URL text content
  4. 04Readability-based main-content extraction
  5. 05Configurable content length

Use cases

  1. 01Find pages for a search query
  2. 02Retrieve readable text from a URL
  3. 03Limit page content for context

MCP support

Exposes DuckDuckGo search and URL-reading tools to MCP clients.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "web": {
      "command": "npx",
      "args": [
        "-y",
        "https://zgagnon.com/webmcp"
      ]
    }
  }
}
Run commandnpx -y https://zgagnon.com/webmcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmwebmcp

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Web MCP Server

A Model Context Protocol (MCP) server that provides web search and content retrieval capabilities for AI assistants.

Features

The server offers two primary tools:

  • search_duckduckgo: Performs web searches using DuckDuckGo and returns URLs of the search results
  • read_url: Fetches and processes content from URLs, removing HTML markup for better readability

Setup

# Install dependencies
npm install

# Development (with auto-reload)
npm run dev

# Production build
npm run build
npm start

Running locally

use

npx https://zgagnon.com/webmcp

or

nix run github:zgagnon.com/webmcp

API

search_duckduckgo

Performs a web search using DuckDuckGo and returns only the URLs of the search results.

Input:

  • query (string, required): The search query to be executed
  • limit (number, optional): Maximum number of results to return (default: 10)

Output:

  • A list of URLs matching the search query
read_url

Fetches and returns the text content of a specified URL, with HTML markup removed. Uses Mozilla's Readability to extract meaningful content.

Input:

  • url (string, required): The URL to fetch content from
  • useReadabilityMode (boolean, optional): Whether to use Mozilla's Readability to extract the main content (default: true)
  • maxLength (number, optional): Maximum length of returned content in characters (default: 10000)

Output:

  • The plain text content from the URL, potentially truncated to fit within context windows

Error Handling

The server will return an error if:

  • Invalid arguments are provided to any tool
  • URL fetching fails
  • Search execution fails

Implementation Details

The server uses:

  • JSDOM and Mozilla's Readability for HTML processing and content extraction
  • Node-fetch for HTTP requests
  • Zod for schema validation

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.