Advertise
MCP.so
Sign In

Fetch MCP Server

@zcaceres

Repository

Basic information

Category

Browser Automation

License

MIT

Transports

stdio

Author

zcaceres

About Fetch MCP Server

A flexible HTTP fetching Model Context Protocol server.

#fetch#web-content#data-fetching

Fetch MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "fetch-mcp-zcaceres": {
      "command": "npx",
      "args": [
        "mcp-fetch",
        "<command>",
        "<url>",
        "[flags]"
      ]
    }
  }
}

Fetch MCP Server tools

5

Perform HTTP requests with full control over method, headers, body, and other fetch options

Get use cases and best practices for using the tools in this MCP server

Execute GraphQL queries and mutations with support for variables and custom headers

Control browsers and pages with Puppeteer - launch/close browsers, open/close pages, execute JavaScript, take screenshots, and more

Manage WebSocket connections - connect, send, receive messages, list connections, and close

Overview

What is Fetch MCP Server?

Fetch MCP Server is an MCP (Model Context Protocol) server that fetches web content and returns it in multiple formats: HTML, JSON, plain text, Markdown, readable article content (via Mozilla Readability), and YouTube video transcripts. It is intended for AI agents and developers who need to programmatically retrieve and process web content.

How to use Fetch MCP Server?

Install with npx mcp-fetch-server or globally via npm install -g mcp-fetch-server. Add it as an MCP server in your client configuration under "mcpServers", or use the CLI: mcp-fetch <command> <url> [flags]. Commands include html, markdown, readable, txt, json, and youtube. All tools accept common parameters: url, headers, max_length, start_index, and proxy.

Key features of Fetch MCP Server

  • Fetch web content as HTML, JSON, plain text, or Markdown
  • Extract article content with Mozilla Readability (strips ads, nav, boilerplate)
  • Extract YouTube video transcripts (via yt-dlp or direct extraction)
  • Proxy support for requests behind firewalls
  • SSRF protection (blocks private/localhost addresses and DNS rebinding)
  • Response size limits and configurable character limit via environment variables

Use cases of Fetch MCP Server

  • An AI agent fetching a webpage to answer user questions about its content
  • Extracting the main article from a news or blog post, free of ads and navigation
  • Retrieving JSON from an API endpoint for structured data processing
  • Obtaining a YouTube video transcript for summarization or translation
  • Browsing web pages with custom headers or through a proxy in restricted networks

Frequently asked questions

1

What content formats does Fetch MCP Server support?

It supports HTML, Markdown, plain text, JSON, readable article content (Markdown via Mozilla Readability), and YouTube transcripts.

2

How can I set a custom character limit for responses?

Use the `max_length` parameter in the tool call, or set the `DEFAULT_LIMIT` environment variable (default 5000 characters; 0 for no limit).

3

Does Fetch MCP Server support proxy requests?

Yes. Pass the `proxy` parameter (e.g., `http://proxy:8080`) to route requests through a proxy.

4

Is there protection against SSRF attacks?

Yes. The server blocks requests to private/localhost addresses and implements DNS rebinding protection.

5

Can I get YouTube transcripts with a specific language?

Yes. Use the `youtube` command or `fetch_youtube_transcript` tool with the optional `lang` parameter (default `"en"`).

Comments

More Browser Automation MCP servers

Fetcher MCP

jae-jae

MCP server for fetch web page content using Playwright headless browser.

1.1K
C

Chrome DevTools for agents

ChromeDevTools

Official Chrome DevTools MCP server from Google that lets coding agents control, debug and profile a live Chrome browser.

53.6K

Fastcrw

us

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary.

286

Screenshot Scout

screenshotscout

Capture screenshots of webpages as images or PDFs with Screenshot Scout.

BrowserTools MCP

AgentDeskAI

Monitor browser logs directly from Cursor and other MCP compatible IDEs.

7.3K
Advertise