Read a brand out of a domain
Building a mockup for a client, a slide with their colours or a landing page in their look starts with the same twenty minutes: open the page, dig through the stylesheet, guess the hex codes, hunt for a logo that is not a screenshot. This does that part.
- 1Enter the domain
- 2Check the source of each value
- 3Copy or download what you need
Try it with: made-with.ai, study-online.io, language-schools.io
Nothing read yet
Type a domain above. Works with any publicly reachable page.
Where each value comes from
The tool shows its source for everything it returns. That matters, because the sources differ in how much they can be trusted.
The logo, in six steps
First the structured data of the page: if a site declares its own logo there, that beats any guess. Then the touch icon, then an SVG in the header, then an image whose name says logo, then the favicon. Last comes the sharing image, and that one often has text on it.
The colours, weighted by area
CSS variables named primary, accent or brand count most, because the site is stating something about itself. After that we measure the actual surfaces: header, navigation, footer, buttons, headline colour, link colour, each weighted by how much of the screen it covers. Anything under ten percent saturation is treated as neutral and becomes a background candidate.
The typeface, honestly
We read the font stack of the headline and the body text. If the family is free to use, you get its name and can load it. If it is licensed, such as Circular or Gotham, we say so and name the closest free equivalent instead of quietly swapping it. We never download or pass on font files.
Questions
Only the address you type reaches our service, which then reads the publicly available start page. The logo, the colours and the typeface information go back to your browser and are processed there. The picture itself is drawn in your browser and never reaches us. We store neither your logo nor the finished image. The address is cached for 24 hours so a repeated request does not open the page again.
Video / Meet Background Generator: Free, nothing stored, cleared for commercial use
Can I use the logo I get here?+
That depends on your rights to the brand, not on this tool. Reading a page is one thing, using a trademark is another. For your own company or a client who has hired you, you are fine.
Why not just take a screenshot?+
A screenshot gives you approximate colours and a pixelated logo. This gives you the exact hex values, and if the site uses an SVG, a logo that stays sharp at any size.
Does it work on every site?+
It works on public pages that render without a login. Sites behind a cookie wall, behind an age check or with aggressive bot protection may return little or nothing. You will see empty fields rather than invented values.
How often can I use it?+
Ten domains per hour and connection. The limit exists so the service does not turn into a free crawler for other people.
Do you store what I look up?+
The result is cached for 24 hours under the domain, so a second request is instant. We do not store the logo, we do not store who asked, and there is no account to tie it to.
Can I call this from my own code?+
The service behind it is a small HTTP endpoint. It is not open to the public today, but if you have a use for it, write to us and tell us what you would build.
Data in, finished file out
This tool reads one page and returns a handful of values. The same approach applied to a product database produces individual brochures, and applied to supplier data it produces catalogues that are ready to print.