crawl-markdown

Use the crawl_markdown tool from the Crawlbase MCP server to fetch clean Markdown from a URL.

When to use

Use this skill when the user wants to read or summarize a webpage's content without dealing with HTML noise. Ideal for articles, documentation, blog posts, and product pages.

Usage

Call the crawl_markdown tool with the url parameter. Optionally pass:

• device: "desktop" (default), "mobile", or "tablet"
• user_agent: custom user agent string
• store: true to push the result to Crawlbase Cloud Storage instead of returning the Markdown. Response then contains rid, stored_at, storage_url, and token_type. Retrieve later with storage_get (use as: "markdown" to convert).

Example prompts

• "Get the markdown content of https://example.com"
• "Summarize the article at https://example.com/article"
• "Extract the main content from https://example.com as markdown"
• "Crawl https://example.com as markdown and store it in Crawlbase Cloud Storage"