What colours and fonts
does this site use?
Paste any URL and read the colours, typefaces and marks it declares in its own code. Every value comes from a declaration you can find by hand in the source, never from sampling an image, so the same site returns the same answer every time.
What it reads
The page itself and up to six stylesheets it links, fetched one at a time. From those: the marks a page nominates for itself (favicon, Apple touch icon, Safari pinned tab, share card), the colours declared in a theme-color tag or in CSS, and the typefaces the site either serves or asks for. Where a value is set through a variable the site never defines, the report names the variable instead of pretending the value is absent.
Questions
Where do the colours come from?
From declarations in the page and its stylesheets: a theme-color meta tag, CSS custom properties whose names say they are brand tokens, and literal colour values. Nothing is sampled from a logo or a screenshot. That matters because sampling is not reproducible - the answer would move with the image resampler and the clustering seed - while a value read from a stylesheet is the same today and next year.
Which one is the brand colour?
We do not say, because we cannot know. Colours are listed in the order they were declared, with the source of each shown, including the custom property name where there is one. A tool that ranked them would be guessing and presenting the guess as a measurement.
Why does it sometimes find no typefaces?
Because some sites set fonts through a CSS variable they never serve in any stylesheet. When that happens the report says so and names the variable, rather than showing an empty list. An empty list and a list we could not finish reading are different facts, and only one of them means the site has no fonts.
Is this a security check?
No. It produces no findings, has no score, and is never part of a scan result. It reads the same public response a browser gets and reports the parts that describe appearance.
Can I point it at a site I do not own?
Yes. It requests only the page and the stylesheets that page links, which is what any visitor's browser already does, and it never probes a path the site did not nominate.