Skip to content
MCP server

mcp.vanshul.com

Public MCP server that reads any web page as clean Markdown for AI agents

Install

mcp.vanshul.com is an open-source MCP server for web search and scraping. It runs as a hosted remote server, so there is nothing to install locally.

Pricing
Open source
Platforms
API, CLI, Self-hosted
Runs
Remotely
Version
1.0.0
Listing updated

Similar MCP servers

Install

Run in a terminal

claude mcp add --transport http mcp https://mcp.vanshul.com/mcp

Add to Settings → Connectors → Add custom connector

https://mcp.vanshul.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp": {
      "url": "https://mcp.vanshul.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mcp": {
      "type": "http",
      "url": "https://mcp.vanshul.com/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "mcp": {
      "serverUrl": "https://mcp.vanshul.com/mcp"
    }
  }
}

Add to Grok Bot desktop app → Connect apps → Add a custom connector

https://mcp.vanshul.com/mcp

Enter a Name (for example "mcp.vanshul.com"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

Full Grok Bot setup guide

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

{
  "mcpServers": {
    "mcp": {
      "type": "http",
      "url": "https://mcp.vanshul.com/mcp"
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

About mcp.vanshul.com

The README describes it as a Cloudflare Worker that gives AI agents read access to live web pages. Each URL is run through Mozilla Readability and Turndown, so navigation, ads and cookie banners are stripped before the text reaches the model.

It exposes four tools over the MCP Streamable HTTP transport: fetch_markdown for the full main content, search_page for only the passages that match a query, fetch_metadata for title, byline, site name and excerpt, and extract_links for outbound links. Clients such as Claude Desktop, Cursor and Continue connect by URL, and stdio-only clients can bridge through mcp-remote.

The server is aimed at developers and agent builders who want current documentation or articles rather than stale training data. Requests are limited to public http(s) addresses, with a 10 second fetch timeout, a roughly 3 MB page cap and a per-IP rate limit. The source is MIT licensed and can be deployed to a Cloudflare account.

Tools this server provides (4)

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

fetch_markdown
Returns the page's main content as clean Markdown, with an optional max_chars limit.
search_page
Returns the passages on a page that match a query, ranked by relevance.
fetch_metadata
Returns title, byline, site name, excerpt and word count for a page.
extract_links
Returns outbound http(s) links and their anchor text from a page.

What it does

Clean Markdown extraction
Converts a page's main content into Markdown, dropping navigation and page chrome.
Query-focused passage search
Returns only the passages matching a query, each with its heading breadcrumb, to save tokens.
Page metadata lookup
Returns title, byline, site name, excerpt and word count for a URL.
Outbound link listing
Lists outbound http(s) links and their anchor text, with an optional limit.
SSRF protection
Blocks private, internal and cloud metadata addresses and re-validates every redirect hop.
No API key required
Public endpoint with no sign-up, and page content is not stored.

Good for

  • Give a coding assistant access to the current version of a documentation page
  • Feed article text into a research or summarisation prompt without boilerplate
  • Pull a specific detail such as a pricing table or config flag from a long page
  • Walk a site's outbound links to find related pages
  • Collect titles and excerpts for citations or link previews

Pricing

Open source

Free public server; source is MIT licensed and self-deployable.

Prices change. Check mcp.vanshul.com before you buy.

mcp.vanshul.com FAQ

Which MCP clients can connect to mcp.vanshul.com?

Any client that supports remote Streamable HTTP MCP servers, such as Claude Desktop, Cursor and Continue. Stdio-only clients can bridge with npx mcp-remote.

Does mcp.vanshul.com need an API key or store page content?

The site says no key is needed and no page content is stored. Extraction is deterministic and uses no LLM.

How does mcp.vanshul.com protect against SSRF?

Only public http(s) URLs are allowed. Localhost, private ranges and cloud metadata addresses are blocked, and each redirect is re-checked.

What tools does mcp.vanshul.com expose?

fetch_markdown, search_page, fetch_metadata and extract_links, as listed in the README's tools table.

Is mcp.vanshul.com free?

Yes. mcp.vanshul.com is open source, so you can use and self-host it for free.

How do I add mcp.vanshul.com to Claude Code?

Run this command in your terminal: claude mcp add --transport http mcp https://mcp.vanshul.com/mcp

What are the best alternatives to mcp.vanshul.com?

The closest alternatives to mcp.vanshul.com are Markfetch, webfetch.run, Fetch URL, WebCrawlerAPI and ToolTrace.

Where this comes from

  • Description: written from the project's README, the project's website and its MCP Registry entry.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers