Skip to content
MCP server

Fetch MCP Server

Fetch web pages as markdown with SSRF protection, reader mode, feeds and robots checks

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI
Runs
Locally
Version
0.8.8
Listing updated
3 days ago

About Fetch MCP Server

Fetch MCP Server is an open-source MCP server for web search and scraping. It runs locally on your machine and is started by your AI client.

Giving a model a raw HTTP client can double as a probe against your internal network. This server keeps SSRF protection on by default and refuses to dial private addresses and cloud metadata endpoints unless an allow_private_hosts setting is changed.

Pages come back in forms that fit a context window: markdown said to be 3-8x smaller than raw HTML, plain text with block structure, or reader mode that isolates the article body. Other tools return head metadata (OpenGraph, JSON-LD, feeds), classified outbound links, parsed sitemaps including gzipped indexes, RSS and Atom entries, and a robots.txt verdict for a path and user agent.

Write methods exist too, with JSON bodies, headers, retries and byte caps. It is MIT-licensed, installs from npm and needs no key.

Install

Run in a terminal

claude mcp add fetch-mcp -- npx -y @yawlabs/fetch-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "fetch-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@yawlabs/fetch-mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "fetch-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@yawlabs/fetch-mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "fetch-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@yawlabs/fetch-mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

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

Add to your client's MCP settings (most use this shape)

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

Tools this server provides (15)

Taken from the project documentation; the running server may differ.

http_get
Bare GET request with headers, auth, timeout, size cap and retry.
http_head
HEAD request with the shared parameter set.
http_options
OPTIONS request with the shared parameter set.
http_post
Write-method HTTP with a JSON or raw body.
http_put
Write-method HTTP with a JSON or raw body.
http_patch
Write-method HTTP with a JSON or raw body.
http_delete
Write-method HTTP with a JSON or raw body.
fetch_html_to_markdown
GET a page and convert it to clean markdown.
fetch_html_to_text
GET a page and convert it to plain text with block structure preserved.
fetch_reader
Reader-mode extraction returning title plus markdown.
fetch_meta
Extract head metadata including OpenGraph, Twitter cards and JSON-LD.
fetch_links
Every outbound link resolved to absolute URLs and classified internal or external.
fetch_sitemap
Parse sitemap.xml including gzipped sitemaps and index chaining.
fetch_feed
Parse an RSS 2.0 or Atom 1.0 feed into entries.
fetch_robots
Return the robots.txt verdict for a path and user agent.

What it does

SSRF guard
Private hosts are blocked by default.
Reader mode
Extracts the article body with title and clean markdown.
Structured parsers
Metadata, links, sitemaps, feeds and robots.txt each have a tool.
Full HTTP verbs
GET, HEAD, OPTIONS, POST, PUT, PATCH and DELETE with auth and timeouts.
Size and retry controls
Byte caps flag truncation and retries honour Retry-After.

Good for

  • Summarise a documentation page without nav clutter
  • Read OpenGraph and JSON-LD for a URL
  • Enumerate a site's pages from its sitemap
  • Check whether a path may be crawled
  • Call a staging webhook and read the response

Pricing

Open source

MIT licence; no key

Prices change. Check yaw.sh before you buy.

Fetch MCP Server FAQ

Is an API key required?

No. The site says there is nothing to provision.

How big are the markdown results?

The tool table says 3-8x smaller than raw HTML.

What licence is it?

MIT, published on npm as @yawlabs/fetch-mcp.

Is Fetch MCP Server free?

Yes. Fetch MCP Server is open source, so you can use and self-host it for free.

How do I add Fetch MCP Server to Claude Code?

Run this command in your terminal: claude mcp add fetch-mcp -- npx -y @yawlabs/fetch-mcp

What are the best alternatives to Fetch MCP Server?

The closest alternatives to Fetch MCP Server are REA, Scrapling MCP Server, Codebase Memory, GitHub and Serena MCP: the IDE for your agent.

Alternatives to Fetch MCP Server

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. MCP server for AI clients to scrape sites with HTTP and stealth browser sessions

    Runs locally
    • 17 tools
    • Open source
    • 104K GitHub stars
    • Web search and scraping
  3. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  4. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  5. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  6. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory