Do AI agents use the canonical URL?
We parse the head of your homepage and record the href of the link rel=canonical element, or its absence.
What it is
A canonical link is one element in the head that names the address you consider authoritative for the page. It resolves the duplicates every site accumulates: www and non-www, trailing slash, tracking parameters, print views.
We read the homepage only, and we report the href verbatim. We do not check that the address resolves or that it matches the URL we fetched.
An absent canonical is a Warning at Low priority, not a Fail. Most sites still work without one, they are just harder to cite.
Why an agent needs it
An agent that summarises your page has to give the reader a link back. Without a canonical it cites whatever address it arrived on, including the one carrying a campaign parameter.
Duplicates also split what an agent understands about you. Two addresses for one page read as two pages with the same content, which is a reason to trust neither.
The Fix
Add `<link rel="canonical" href="https://yourcompany.com/">` to the head of the homepage, and the equivalent absolute address on every other page.
Use one absolute URL with the scheme and hostname you actually serve. A relative or protocol-relative href defeats the purpose.
Make sure the canonical of a page points at itself, not at the homepage. A whole site canonicalised to its root is a common template mistake.
This one is a change in your own markup or edge configuration, so the Fix Pack cannot write it for you. A Report names every page it affects, with the Evidence.