<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport">
  <title>My Website</title>
  <link rel="stylesheet">
  <script src="app.js"></script>
</head>
<body>
  <div class="container">
    <header class="header">
      <nav class="navbar">
        <ul class="nav-list">
          <li><a href="/">Home</a></li>
          <li><a href="/about">About</a></li>
          <li><a href="/contact">Contact</a></li>
        </ul>
      </nav>
    </header>
    <main class="content">
      <article class="post">
        <h1 class="title">Welcome</h1>
        <p class="lead">
          This is the main content...
        </p>
        <div class="features">
          <div class="feature-card">
            <h3>Feature One</h3>
            <p>Description here...</p>
          </div>
          <div class="feature-card">
            <h3>Feature Two</h3>
            <p>Another description...</p>
          </div>
        </div>
      </article>
      <aside class="sidebar">
        <div class="widget">
          <h4>Related Posts</h4>
          <ul>
            <li><a href="#">Post 1</a></li>
            <li><a href="#">Post 2</a></li>
          </ul>
        </div>
      </aside>
    </main>
    <footer class="footer">
      <div class="footer-content">
        <p>&copy; 2024 Company</p>
        <nav class="footer-nav">
          <a href="/privacy">Privacy</a>
          <a href="/terms">Terms</a>
        </nav>
      </div>
    </footer>
  </div>
  <script>
    document.addEventListener('DOMContentLoaded', function() {
      console.log('Page loaded');
      initializeApp();
    });
  </script>
</body>
</html>
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <title>Another Page</title>
</head>
<body>
  <div id="app">
    <header>
      <h1>Dashboard</h1>
    </header>
    <main>
      <section class="stats">
        <div class="stat-card">
          <span>Users</span>
          <strong>1,234</strong>
        </div>
      </section>
    </main>
  </div>
</body>
</html>

# Welcome

This is the main content...

## Features

### Feature One

Description here...

### Feature Two

Another description...

---

## Getting Started

Just fetch any URL and get
clean, AI-ready Markdown.

### Installation

```bash
curl https://api.mdfetch.ai/URL
```

### Usage

Simple HTTP API call returns
clean Markdown content.

## Why MdFetch?

- Clean output
- No HTML junk
- No scripts
- No ads
- AI-ready

## Documentation

### API Reference

GET /https://example.com

Returns Markdown content.

### Authentication

Use Bearer token in header.

### Rate Limits

10 free requests per day.
Paid plans available.

---

## Examples

### Basic Fetch

Fetch any webpage:

```
GET /https://docs.example.com
```

### With Options

Add parameters for control.

## Support

Contact us for help.

---

# Another Document

## Overview

More content here...

### Section One

Details about section one.

### Section Two

Details about section two.

## Conclusion

Summary of everything.

Turn any webpage into
AI-ready Markdown

Works with Claude Code, Cursor & any MCP client.
Save up to 90% on tokens.

Continue with GitHub Continue with Google

No credit card required

Try it free — no signup required

Save AI Tokens

Clean Markdown = fewer tokens. No HTML bloat, scripts, or ads.

Smart Content

Requests raw source from GitHub/GitLab or native Markdown when available.

API & MCP Ready

Simple HTTP API. Works with any MCP client like Claude Code and Cursor.

Works with

+ any MCP-enabled tool

Simple, transparent pricing

No credit card required to start

$0.0005

per request

Pay per use, billed every 2 weeks

10 free requests/day to get started

Estimate your cost

requests

$0.3000 total cost

Start now

Large usage? Get volume discounts

We offer discounted pricing for high-volume workloads. Perfect for production AI agents, large-scale RAG pipelines, and research teams.

Install and Usage

Terminal

curl https://api.mdfetch.ai/https://github.com/user/repo \
  -H "Authorization: Bearer YOUR_TOKEN"

curl https://api.mdfetch.ai/https://github.com/user/repo \
  -H "Authorization: Bearer YOUR_TOKEN"

Terminal

claude mcp add --transport http mdfetch https://mcp.mdfetch.ai \
  --header "Authorization: Bearer YOUR_TOKEN"

claude mcp add --transport http mdfetch https://mcp.mdfetch.ai \
  --header "Authorization: Bearer YOUR_TOKEN"

mcp_settings.json

{
  "mcpServers": {
    "mdfetch": {
      "url": "https://mcp.mdfetch.ai",
      "headers": {
        "Authorization": "Bearer YOUR_TOKEN"
      }
    }
  }
}

{
  "mcpServers": {
    "mdfetch": {
      "url": "https://mcp.mdfetch.ai",
      "headers": {
        "Authorization": "Bearer YOUR_TOKEN"
      }
    }
  }
}

See the full documentation for more details.

Frequently Asked Questions

What is MdFetch?

MdFetch converts web pages to clean, AI-ready Markdown via HTTP API or native MCP server. It uses content negotiation to fetch raw Markdown from GitHub, GitLab, and documentation sites when available, reducing token usage by up to 90%.

How much does MdFetch cost?

MdFetch charges $0.0005 per request with pay-as-you-go pricing. Credits never expire. A limited free tier is available for testing and development.

Does MdFetch work with Claude Code?

Yes, MdFetch provides a native MCP server that integrates directly with Claude Code, Cursor, Windsurf, and other MCP-compatible AI coding assistants. Installation takes less than a minute.

What is content negotiation in MdFetch?

Content negotiation automatically fetches source Markdown or raw code from platforms like GitHub and GitLab instead of rendered HTML. This provides cleaner data, reduces token usage, and improves LLM response quality.

What formats does MdFetch support?

MdFetch returns clean Markdown optimized for LLMs. It automatically handles content negotiation for GitHub, GitLab, and documentation sites, falling back to HTML-to-Markdown conversion for other sites.

Is MdFetch suitable for production use?

Yes, MdFetch is built for production with reliable infrastructure, rate limiting, and transparent usage pricing. It's designed for AI agents, RAG pipelines, and automated workflows.

More questions? Check the documentation or contact us.

Turn any webpage into AI-ready Markdown