Where does this link really go?

A link can display one thing and contact another. Paste it here and you get the host your browser will actually connect to, plus the specific technique used to disguise it. Nothing is sent anywhere: the parsing happens in this tab, using the browser's own URL parser rather than a guess.

The tricks this catches

The @ trick. Everything before an @ in a URL is a username, not a destination. https://paypal.com@evil.example/ contacts evil.example, and the familiar name is decoration. This is the one that fools people who were told to "check the link before clicking".

Brand in a subdomain. paypal.com.security-check.example.net ends in example.net. Only the labels at the end decide where the request goes; anything to the left is chosen freely by whoever owns that domain.

Punycode hosts. A host starting xn-- is an internationalised name encoded back into ASCII. Some of those render as characters that look exactly like Latin letters, which is how a convincing lookalike domain is built.

Nested URLs. A second URL sitting in a query parameter usually means a redirect: the first host is a hop, and the destination is the one inside.

Raw IP addresses and odd ports. Neither is proof of anything, but a legitimate consumer service almost never asks you to visit an IP address on a non-standard port.

What it does not tell you

This decodes structure, not intent. A perfectly ordinary-looking host can be malicious, and a flagged one can be harmless. What it removes is the specific class of deception where the link is built to read as one thing and resolve as another.

One honest limit on the registrable-domain check: multi-part suffixes such as co.uk or com.au mean the last two labels are not always the domain. A short list of common ones is handled here, and it is not exhaustive.