What does a URL to Markdown API not do?
It does not run JavaScript, log in, click, or crawl a site. It reads one public HTTPS page and returns clean Markdown.
No JavaScript, no browser
ScriptMasterLabs' reader fetches the raw page source directly. Scripts on the page never execute. Content loaded later by scripts stays missing. Pages built only by JavaScript return thin results.
No logins, no clicks, no cookies
It cannot open pages behind a login. It enters no password and holds no account. It does not click buttons or submit forms. It does not scroll, wait, or accept cookies.
One page, not a whole site
Each call reads exactly one public HTTPS address. It does not follow links across a site. It does not map, spider, or archive domains. Private, local, or blocked addresses are refused. Failed or refused reads cost nothing.
What a real job showed?
ScriptMasterLabs measured eight public documentation pages on October 28, 2026. Returned Markdown was 95.3 percent smaller than raw HTML. The eight reads returned 129,881 characters of Markdown. The same pages carried 2,773,841 bytes of HTML. Two attempted pages were refused with no charge. One address returned 404, so nothing was delivered. The other page was 1.2 megabytes in one file. It returned the no-charge result, not partial Markdown.
Use a different tool when
Use a rendering crawler for JavaScript-heavy pages. Firecrawl and Jina Reader render pages this reader cannot. Use a browser tool for logins and clicks. The wrong job is how a good product gets a bad reputation.
What it does instead?
It turns one known public page into clean Markdown. ScriptMasterLabs charges 0.001 USDC per accepted read. Failed reads cost 0 USDC on Base. A free preview needs no key or wallet. For that one job, it beats a heavier crawler.
See if your page fits
Check your page type against these limits first. Then see the reader that fits this job before paying anything.
Common questions
Can it read a page behind a login? No, it cannot read pages behind logins.
Does a failed read cost money? No, failed reads cost 0 USDC.
What should I use for JavaScript pages? Use a rendering crawler like Firecrawl or Jina Reader.