Runs the same UI tasks through model outputs and measures whether a coding agent follows a design system it did not invent.
Open source
Open-source tools.
One loop for AI‑native design systems: ds‑context prepares context, ui‑repair repairs code, ds‑eval measures whether the agent follows the system, prompt‑regress compares prompt versions. Specification and production checks follow.
Builds a short model pack from tokens, components, and a spec, and prices each packing strategy in tokens.
<Button variant="primary" />
Replaces raw controls, hex, and a homemade modal with system components, and shows which checks flipped.
Compares two prompt versions on one UI suite: what improved, what broke, instead of an average score.
Extracts variables, styles, and components from Figma into a markdown spec an agent can read.
Finds hardcoded colors, spacing, and fonts in code and suggests a token instead.
Measures how much of the codebase actually imports the design system, and where components were rebuilt locally.
Paste a live URL and get a report on how many colors, type styles, spacings, and radii the page actually uses.