curl.md

Fetch any URL as clean markdown.

When to Use

Use curl.md when:

• The user asks to read, summarize, or extract content from a web page
• You need to fetch documentation before answering a technical question
• The user shares a URL and asks about its content
• You need to research a topic using a specific web source
• The user asks “what does this page say about X”

Usage

# Fetch a URL as markdown
curl curl.md/<url>

# Focus output with a query (AI-powered extraction)
curl curl.md/<url>?q=<query>

# Bypass cache for fresh content
curl curl.md/<url>?fresh=

Parameters

No protocol needed — curl curl.md/example.com automatically uses HTTPS.

Examples

# Get React docs as markdown
curl curl.md/react.dev

# Get focused content
curl curl.md/react.dev?q=fullstack+support

# Wikipedia article section
curl curl.md/en.wikipedia.org/wiki/Linux?q=kernel+history

# GitHub REST API docs
curl curl.md/docs.github.com/en/rest?q=rate+limiting

# Bypass cache for a fast-changing page
curl curl.md/status.example.com?fresh

Tips

• Use ?q= to reduce token usage. Without it, you get the entire page. With it, only relevant sections are returned verbatim.
• Pipe to tools for further processing: curl curl.md/example.com | head -100
• Known doc sites (like Cloudflare, Vercel, Bun, PlanetScale) are fetched as native markdown for higher quality output.
• Content is cached for 15 minutes. Use ?fresh if you need the latest version.

Limitations

• Pages that require JavaScript rendering (SPAs without SSR) may return incomplete content.
• Paywalled or authenticated content is not accessible.
• Very large pages are truncated to ~100k characters when using ?q=.
• Returns JSON { "error": "..." } if the upstream site returns an error.

MCP Server

An MCP server is also available with a fetch_page tool:

npx add-mcp curl.md/mcp

fetch_page Tool