🤖 DeepCrawl AI — Universal Website to Markdown Crawler Pro (Model Context Protocol Server)

A high-performance Model Context Protocol (MCP) Server and deep domain crawler that turns any website, documentation hub, or sitemap into clean, LLM-ready Markdown for Claude Desktop, Cursor IDE, LangChain, RAG pipelines, and Autonomous AI Agents.

⚡ Overview

DeepCrawl AI is a lightweight, high-speed alternative to Firecrawl and Crawl4AI. It recursively crawls documentation portals (e.g. Stripe, FastAPI, Supabase) and blogs, strips all web noise (ads, cookie banners, tracking scripts, navigation menus), and formats the core content into token-budgeted GitHub-Flavored Markdown.

• 🛡️ Zero Noise HTML Cleaner: Strips header/footer boilerplates, cookie banners, popups, and sidebar ads.

• 🌲 BFS Deep Domain Crawler: Recursively discovers internal links with strict domain boundary locking.

• 🗺️ Fast Sitemap & URL Mapper: Maps all available URLs across an entire domain in seconds.

• 🔌 Zero-Config MCP Integration: Plug-and-play with Claude Desktop, Cursor, and Antigravity agents.

Related MCP server: Spider MCP Server

🛠️ MCP Tools & Capabilities

The server exposes the following Model Context Protocol (MCP) tools:

🔌 Quickstart: Connect to Claude Desktop & Cursor

1. Claude Desktop Configuration

Add this server to your claude_desktop_config.json:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

• Windows: %APPDATA%\Claude\claude_desktop_config.json

• Linux: ~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "deep-markdown-crawler": {
      "command": "python3",
      "args": [
        "/path/to/deep-markdown-crawler-api/mcp_server.py"
      ]
    }
  }
}

2. Cursor IDE Configuration (.cursor/mcp.json)

{
  "mcpServers": {
    "deep-markdown-crawler": {
      "command": "python3",
      "args": ["/path/to/deep-markdown-crawler-api/mcp_server.py"]
    }
  }
}

📦 Python SDK & CLI Usage

pip install deep-markdown-crawler

from deep_markdown_crawler import scrape_url_to_markdown, crawl_domain_to_markdown

# 1. Scrape Single URL
doc = scrape_url_to_markdown("https://fastapi.tiangolo.com/tutorial/")
print(f"Title: {doc['title']} | Tokens: ~{doc['token_count']}")
print(doc['markdown'][:500])

# 2. Deep Crawl Entire Domain
crawl = crawl_domain_to_markdown(
    start_url="https://fastapi.tiangolo.com/tutorial/",
    max_pages=5,
    max_depth=2
)
print(f"Crawled {crawl['total_pages_crawled']} pages ({crawl['total_tokens_estimated']} tokens).")

🌐 Serverless REST API Endpoints

📄 License

MIT License. Created by Meanus Arcanus.