
Video to Code turns a video of a user interface into a build spec that your AI Code (or any coding agent) implements.
Share a screen recording, a design shot or an app demo, and it returns the screens, the layout of each one, components, colours as hex values, the on-screen copy word for word, and a motion table: every animation with its element, what it does (moves, scales, fades, turns, changes colour), its delay, duration and easing as a CSS curve, and from → to values ready for CSS or SwiftUI. For video files (an upload or a direct link), motion is measured from the frames themselves: starts and durations to about ±25 ms on 60 fps video (less exact at lower frame rates), the direction and distance of each move, scale and fades, plus key frames as images so the agent sees exact colours.
https://videotocode.app/api/mcpNone in the store listings. The server needs sign-in, so we could not check it