# Readmd

> Readmd is a paid MCP server for web search and scraping. Turn any web page into clean Markdown with an outline and token counts, per page. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API
- **Website:** https://readmd.canisell.xyz
- **MCP Registry name:** xyz.canisell/readmd
- **Categories:** web search and scraping
- **Listing updated:** 2026-11-15

## About

Readmd fetches a URL over plain HTTP, drops navigation, footers, sidebars and cookie banners, and returns GitHub-flavoured Markdown with headings, lists, tables, code blocks and absolute links. Alongside come metadata such as title, author, dates and canonical URL, an outline of headings, a word count and a token estimate.

Options include a full-page mode, dropping links to save tokens, returning images, capping tokens, and splitting the text into RAG-ready chunks with heading paths. It also converts native Markdown, plain text, JSON and RSS or Atom feeds. It respects robots.txt under the user-agent token readmd, caps pages at 3 MB, re-checks redirects and refuses private addresses.

There is no headless browser, so JavaScript-only pages, PDFs and images are not supported. A page costs $0.001 in USDC on Base via x402, or $0.005 for up to ten, and failures are never charged. A free preview shows metadata, outline and the first 500 characters.

## Features

- **Main-content extraction:** Removes page chrome and keeps the article as Markdown.
- **Metadata and outline:** Title, author, dates, heading outline and token estimate.
- **RAG chunks:** Optional chunks with heading paths.
- **Batch reads:** Up to ten pages in one call.
- **Polite fetching:** Respects robots.txt, caps size and refuses private addresses.
- **Failures are free:** Only successfully read pages are charged.

## Good for

- Feed an article into a prompt without page clutter
- Prepare documentation pages for a RAG index
- Estimate token cost before reading a page
- Convert an RSS feed to Markdown

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http readmd https://readmd.canisell.xyz/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://readmd.canisell.xyz/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "readmd": {
      "url": "https://readmd.canisell.xyz/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "readmd": {
      "type": "http",
      "url": "https://readmd.canisell.xyz/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "readmd": {
      "serverUrl": "https://readmd.canisell.xyz/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "readmd": {
      "type": "http",
      "url": "https://readmd.canisell.xyz/mcp"
    }
  }
}
```

## Tools

- `read_url`: Read one web page as clean Markdown
- `read_urls`: Read up to 10 web pages as Markdown
- `preview_url`: Free preview of a page's metadata, outline and first 500 characters

## FAQ

### What does it cost?

$0.001 per page or $0.005 for up to 10 pages, paid in USDC via x402 on Base.

### Is there a free option?

A free preview returns metadata, outline, counts and the first 500 characters, rate-limited per IP.

### Does it run JavaScript?

No. Pages that need JavaScript to render are not supported.

### Does it obey robots.txt?

Yes, using the user-agent token readmd.

### Which tools does it expose?

read_url, read_urls and preview_url.

### Is Readmd free?

No. Readmd is a paid product.

### How do I add Readmd to Claude Code?

Run this command in your terminal: claude mcp add --transport http readmd https://readmd.canisell.xyz/mcp

### What are the best alternatives to Readmd?

The closest alternatives to Readmd are Scrapling MCP Server, Firecrawl MCP Server, SearXNG Search, NeuralVerge and AI-Powered directory.

## Alternatives

- [Scrapling MCP Server](https://ai-powered.com/mcp/scrapling-mcp-server): Scrapling MCP Server is an open-source MCP server for web search and scraping.
- [Firecrawl MCP Server](https://ai-powered.com/mcp/firecrawl-mcp-server): Firecrawl MCP Server is a freemium MCP server for web search and scraping.
- [SearXNG Search](https://ai-powered.com/mcp/searxng-search): SearXNG Search is an open-source MCP server for web search and scraping.
- [NeuralVerge](https://ai-powered.com/mcp/neuralverge): NeuralVerge is an MCP server for web search and scraping.
- [AI-Powered directory](https://ai-powered.com/mcp/ai-powered-directory): AI-Powered directory is a free MCP server for developer tools.
- [Shopify MCP Server (any store)](https://ai-powered.com/mcp/shopify-mcp-server-any-store): Shopify MCP Server (any store) is a paid MCP server for e-commerce.
- [Chanroll: YouTube Channel Video Scraper (Official API)](https://ai-powered.com/mcp/chanroll-youtube-channel-video-scraper-official-api): Chanroll: YouTube Channel Video Scraper (Official API) is a paid MCP server for design and media.
- [Vidvett: YouTube Video Details Scraper (Official API)](https://ai-powered.com/mcp/vidvett-youtube-video-details-scraper-official-api): Vidvett: YouTube Video Details Scraper (Official API) is a paid MCP server for design and media.

---
Source: https://ai-powered.com/mcp/readmd
