Fetch MCP server

Fetches and processes web content into markdown with intelligent image handling for seamless integration into research workflows.
Back to servers
Provider
Jeremy Nixon
Release date
Mar 18, 2025
Language
TypeScript

The MCP-Fetch server is a Model Context Protocol server designed to help Claude Desktop (or other MCP clients) fetch web content and process images appropriately. It allows Claude to access online information and handle images within its context by managing downloads, transformations, and clipboard operations.

Installation

Quick Installation for Claude Desktop

Add the following to your Claude Desktop configuration file located at ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "tools": {
    "fetch": {
      "command": "npx",
      "args": ["-y", "@kazuph/mcp-fetch"]
    }
  }
}

This configuration automatically downloads and runs the latest version of the tool when needed.

Installing via Smithery

You can install MCP Fetch for Claude Desktop automatically using Smithery:

npx -y @smithery/cli install @kazuph/mcp-fetch --client claude

Required System Setup

For the tool to work properly, you need to enable Accessibility permissions for Claude:

  1. Open System Settings
  2. Go to Privacy & Security > Accessibility
  3. Click the "+" button
  4. Add Claude from your Applications folder
  5. Turn ON the toggle for Claude

This accessibility setting is required for automated clipboard operations to work correctly.

Usage

Finding the Claude Desktop Configuration

You can easily access the Claude Desktop configuration through the app:

  1. Open Claude Desktop
  2. Click Claude on the Mac menu bar
  3. Click "Settings"
  4. Click "Developer"

The configuration file is located at: ~/Library/Application Support/Claude/claude_desktop_config.json

Using the Fetch Tool

Once configured, you can use the fetch capability directly in your Claude Desktop conversations. The tool will:

  • Retrieve web content from URLs you provide
  • Automatically convert webpage content to markdown
  • Process and prepare images for optimal use with Claude
  • Handle clipboard operations for seamless content transfer

Image Processing Details

When processing images from web content, MCP-Fetch applies these limits:

  • Maximum 6 images per group
  • Maximum height of 8000 pixels per group
  • Maximum size of 30MB per group

If content exceeds these limits, images will be automatically split into multiple groups, and you'll need to paste (Cmd+V) multiple times to include all content.

Technical Notes

  • This tool is designed for macOS only due to its dependency on macOS-specific clipboard operations
  • The server automatically handles animated GIFs by extracting their first frame
  • When multiple images are found, they are merged vertically with consideration for size limits

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.

Want to 10x your AI skills?

Get a free account and learn to code + market your apps using AI (with or without vibes!).

Nah, maybe later