Fetch MCP Server

What is Fetch MCP Server about?

Fetch MCP Server provides a set of tools for retrieving web resources and converting them into convenient data formats. It supports fetching raw HTML, parsing JSON endpoints, extracting plain text from pages, and converting HTML to Markdown, all via the Model Context Protocol.

How to use Fetch MCP Server?

1. Clone the repository and install dependencies:

   git clone https://github.com/zcaceres/fetch-mcp.git
   cd fetch-mcp
   npm install
   

2. Build the TypeScript source:

   npm run build
   

3. Start the server (STDIO mode):

   npm start
   

4. Invoke the desired tool (e.g., fetch_html) by sending an MCP request with the required url and optional headers fields.

Key features of Fetch MCP Server

• Modern fetch API for HTTP requests
• Custom request headers support
• Four content‑type tools: fetch_html, fetch_json, fetch_txt, fetch_markdown
• HTML parsing via JSDOM
• HTML‑to‑Markdown conversion using TurndownService
• TypeScript codebase with watch‑mode development and test suite

Use cases of Fetch MCP Server

• Pulling article bodies for summarisation or analysis
• Retrieving API responses in JSON for downstream processing
• Converting web pages to Markdown for documentation or content migration
• Extracting clean plain‑text for natural‑language processing pipelines

FAQ from the Fetch MCP Server

Q: Do I need to run a separate server process? A: The server runs on standard input/output; you can start it with npm start and interact via MCP.

Q: Can I add my own headers (e.g., authentication tokens)? A: Yes, each tool accepts an optional headers object that will be merged into the request.

Q: Is there a persistent state or storage? A: No, the server is stateless and only fetches/transform content on demand.

Q: How do I integrate the server with a desktop application? A: Add the following configuration to your app’s MCP server list, pointing to the built dist/index.js file:

{
  "mcpServers": {
    "fetch": {
      "command": "node",
      "args": ["{ABSOLUTE PATH TO FILE HERE}/dist/index.js"]
    }
  }
}

Fetch MCP Server

This MCP server provides functionality to fetch web content in various formats, including HTML, JSON, plain text, and Markdown.

Components

Tools

• fetch_html

  • Fetch a website and return the content as HTML
  • Input:
    • url (string, required): URL of the website to fetch
    • headers (object, optional): Custom headers to include in the request
  • Returns the raw HTML content of the webpage

• fetch_json

  • Fetch a JSON file from a URL
  • Input:
    • url (string, required): URL of the JSON to fetch
    • headers (object, optional): Custom headers to include in the request
  • Returns the parsed JSON content

• fetch_txt

  • Fetch a website and return the content as plain text (no HTML)
  • Input:
    • url (string, required): URL of the website to fetch
    • headers (object, optional): Custom headers to include in the request
  • Returns the text content of the webpage with HTML tags, scripts, and styles removed

• fetch_markdown

  • Fetch a website and return the content as Markdown
  • Input:
    • url (string, required): URL of the website to fetch
    • headers (object, optional): Custom headers to include in the request
  • Returns the content of the webpage converted to Markdown format

Resources

This server does not provide any persistent resources. It's designed to fetch and transform web content on demand.

Getting started

1. Clone the repository
2. Install dependencies: npm install
3. Build the server: npm run build

Usage

To use the server, you can run it directly:

npm start

This will start the Fetch MCP Server running on stdio.

Usage with Desktop App

To integrate this server with a desktop app, add the following to your app's server configuration:

{
  "mcpServers": {
    "fetch": {
      "command": "node",
      "args": [
        "{ABSOLUTE PATH TO FILE HERE}/dist/index.js"
      ]
    }
  }
}

Features

• Fetches web content using modern fetch API
• Supports custom headers for requests
• Provides content in multiple formats: HTML, JSON, plain text, and Markdown
• Uses JSDOM for HTML parsing and text extraction
• Uses TurndownService for HTML to Markdown conversion

Development

• Run npm run dev to start the TypeScript compiler in watch mode
• Use npm test to run the test suite

License

This project is licensed under the MIT License.