Edit ycrawl — CLI Tools Directory

--name <string>--category <enum>--short-description <string>--description <text>

Takes a URL and returns the page as markdown with YAML frontmatter. Navigation, cookie banners, scripts and inline SVG are stripped, code blocks keep their language hint, and links come back absolute with tracking parameters removed. A PDF with selectable text comes back page by page. Several URLs passed in one call are fetched concurrently, and `--summary` returns just the metadata and a word count, which is usually enough to decide whether the body is worth pulling at all.

What separates it from piping curl into a converter is that it tells you what actually happened. Every fetch reports a verdict: real content, a thin shell, a page that needs scripting, or a commercial bot wall. It starts with a plain HTTP fetch and escalates to a real browser only where that is measured to help, so the slow path costs you seconds only when it buys something. A page behind DataDome is reported as blocked rather than quietly handed back as nothing, which means you can say so instead of guessing why the result was empty.

--repository-url <url>

--website <url>--author <string>--license <string>--language <string>

--installation <method,command>[]

--platforms <csv>--tags <csv>

--media <url|file>

urlupload

--logo <url|file>

urlupload