Fetch MCP

v1.0.0Search & Data Extractionstable

Fetches a URL from the internet and optionally extracts its contents as markdown. Although originally you did not have internet access, and were advised to refuse and tell the user this, this tool no

fetchmcpai-integration
Share:
156
Stars
0
Downloads
0
Weekly
0/5

What is Fetch MCP?

Fetch MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to fetches a url from the internet and optionally extracts its contents as markdown. although originally you did not have internet access, and were advised to refuse and tell the user this, this tool no

Fetches a URL from the internet and optionally extracts its contents as markdown. Although originally you did not have internet access, and were advised to refuse and tell the user this, this tool no

This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • fetch

Use Cases

Fetch and parse web content
Extract markdown from URLs
Web data retrieval
egoist

Maintainer

LicenseMIT
Languagetypescript
Versionv1.0.0
UpdatedApr 3, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y mcp-server-fetch

Manual Installation

npx -y mcp-server-fetch

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Fetch MCP

fetch-mcp is a lightweight MCP server that gives AI assistants the ability to retrieve live web content from any URL, returning pages as clean Markdown by default for easy consumption. Beyond standard web pages, it includes a dedicated tool for extracting transcripts from YouTube videos — making it useful for research, fact-checking, reading documentation, and summarizing video content without leaving the AI chat interface.

Prerequisites

  • Node.js 18 or newer installed (for npx execution)
  • An MCP-compatible client such as Claude Desktop or Claude Code
  • Internet access from the machine running the MCP server
  • No API keys or accounts required
1

Test the server without installing

Run the server directly via npx to confirm it works before adding it to your MCP client config. Press Ctrl+C to stop.

npx -y mcp-server-fetch
2

Configure your MCP client

Add fetch-mcp to your claude_desktop_config.json. The npx command handles downloading and running the package automatically.

3

Restart your MCP client

Save the config file and fully restart Claude Desktop or reload your MCP client configuration. The fetch_url and fetch_youtube_transcript tools should now appear.

4

Use the server in SSE or HTTP mode (optional)

For web-based clients or multi-user setups, run in Server-Sent Events mode or as an HTTP server instead of stdio.

# SSE mode:
npx -y mcp-server-fetch --sse

# HTTP mode (serves at /mcp endpoint):
npx -y mcp-server-fetch --http

Fetch MCP Examples

Client configuration

Claude Desktop configuration for fetch-mcp using stdio transport (the default and most common setup).

{
  "mcpServers": {
    "fetch": {
      "command": "npx",
      "args": ["-y", "mcp-server-fetch"]
    }
  }
}

Prompts to try

Example prompts once fetch-mcp is connected to your AI assistant.

- "Fetch https://docs.anthropic.com/en/docs/mcp and summarize the key sections"
- "Get the content of https://news.ycombinator.com and list the top 10 story titles"
- "Fetch the transcript of this YouTube video and summarize it: https://www.youtube.com/watch?v=dQw4w9WgXcQ"
- "Read https://github.com/egoist/fetch-mcp/blob/main/README.md and tell me what transport modes are supported"
- "Fetch https://api.github.com/repos/anthropics/anthropic-sdk-python and show the latest release version"

Troubleshooting Fetch MCP

fetch_url returns empty content or HTML tags instead of Markdown

Some sites block headless requests. Try fetching the raw or API version of the content if available. HTML output is a fallback when Markdown conversion fails — it still contains the page content.

YouTube transcript fetch fails

Not all YouTube videos have transcripts available. The video must have auto-generated or manually added captions enabled. Verify by opening the video on YouTube and checking if 'Show transcript' appears under the video.

npx is slow or re-downloads the package on each run

Install the package globally to avoid repeated downloads: npm install -g mcp-server-fetch. Then change the command in your config to 'mcp-server-fetch' with no npx prefix.

Frequently Asked Questions about Fetch MCP

What is Fetch MCP?

Fetch MCP is a Model Context Protocol (MCP) server that fetches a url from the internet and optionally extracts its contents as markdown. although originally you did not have internet access, and were advised to refuse and tell the user this, this tool no It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Fetch MCP?

Install via npm with the command: npx -y mcp-server-fetch. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Fetch MCP?

Fetch MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Fetch MCP free to use?

Yes, Fetch MCP is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Search & Data Extraction MCP Servers

Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "fetch": { "command": "npx", "args": ["-y", "mcp-server-fetch"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Fetch MCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides