About our crawler

Mailsound checks the links and images in email HTML that a user submits to the tool. Each request identifies itself with the User-Agent:

Mailsound/1.0 (+https://mailsound.virtunet.io/about-our-crawler)

What it fetches

Only URLs that appear as href or image src values in HTML a user pasted or uploaded, one HEAD (then GET if needed) request per URL, up to 5 redirect hops, at most 2 MB of response body read. No credentials, no custom headers, no request body.

How a host is resolved

Before each request we resolve the hostname over DNS and refuse private, loopback, link-local and Cloudflare-internal addresses; the request itself is then made by hostname, so a name whose DNS answer changes between the lookup and the request could still be fetched at a different address. The Workers fetch API gives us no way to pin the connection to the address we checked.

robots.txt

robots.txt is not consulted for these checks: each one is a one-off link check requested by the person who submitted the email HTML, not a crawl of your site.

How to block it

Block requests carrying the User-Agent above, or write to support@mailsound.virtunet.io and we will stop checking a specific host.