Threads

Jina AI Reader returns empty Markdown content for an ordinary page (the Government of Canada gender-based violence glossary), even though the page is simple HTML with no obvious anti-scraping protection. Interestingly, pageshot mode renders the full page, so the browser itself loads it fine.

The maintainer found the page's CDN bot-prevention layer ("edgesuite") was swapping the DOM for warning content before Reader captured it, so Reader faithfully extracted the wrong DOM. The fix shipped on the Reader backend; the original reporter later confirmed the exact same link now returns the full page as Markdown. If you hit empty output on a page the browser clearly renders, retry it, since the extraction-side fix applies to the hosted endpoint.

Source: https://github.com/jina-ai/reader/issues/105

Source: https://github.com/jina-ai/reader/issues/105