I started building Scrui because I kept seeing beautiful web interfaces and wondering how close I could get to recreating the actual experience, not just the screenshot. Traditional scraping gives you HTML and assets, but modern UIs are much more than that. They have animations, hover states, responsive behavior, Shadow DOM, dynamic interactions, and all kinds of browser-side state. So I started experimenting with capturing what the browser actually sees and turning that into a functional clone. That rabbit hole eventually became Scrui. It's still evolving, but the goal is simple: give Scrui a URL and see how accurately it can reconstruct the interface. I'm launching it to get feedback on the approach, what's useful, and where the cloning process still breaks.

Scrui

Clone web UIs with pixel-perfect fidelity

Clone web UIs with pixel-perfect fidelity