Skip to content
Cloudhal

Cloudhal-Fetch

You found Cloudhal-Fetch/1 (+https://cloudhal.dev/fetch) in your logs. It is the media fetcher of Cloudhal, an open-source site platform.

Who sends it

Cloudhal is self-hosted: every site runs on its owner’s own Cloudflare account or server. A request with this user agent comes from one such site, when its owner or an agent they approved imports an image or file by URL. The Cloudhal project does not crawl the web and does not run these requests itself.

What it requests

  • A single GET for the exact URL someone gave it. It does not follow links or crawl.
  • No cookies, no Authorization header, no headers forwarded from the person’s browser.
  • HTTPS on port 443 only; at most 3 redirects, each checked again.
  • Media types only (images, video, audio, documents). Web pages are refused.
  • At most 25 MB, within 30 seconds.

Blocking it

Because each request comes from a different site’s infrastructure, block by user agent: refuse requests whose User-Agent starts with Cloudhal-Fetch/. The fetcher treats any status other than 200 as a failure and tells the person the import did not work.

Contact

If a Cloudhal site is misusing the fetcher, write to security@cloudhal.com with the request time, URL and source address. The rules above are specified in RFC-0004 §5.7.