design-system-extract

Crawl public websites and extract design tokens into a DESIGNS bundle.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/sujithatzackriya/skills --skill design-system-extract
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-system-extract
Source: https://github.com/sujithatzackriya/skills/tree/main/skills/design-system-extract
Command: npx skills add https://github.com/sujithatzackriya/skills --skill design-system-extract

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, mime-types, node:fs/promises, node:path, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill turns a public website’s visual design into a reusable design system bundle, so you don’t have to manually infer colors, typography, spacing, and components.

Core Features & Use Cases

  • Live multi-page crawling and deterministic extraction: visits a site’s homepage and discovered internal links (nav/footer/body) and captures rendered HTML, computed styles, screenshots, and referenced assets.
  • Token and CSS bundle generation: emits a four-file design-system bundle into DESIGNS/<hostname>/ containing a narrative reference, DTCG-style tokens JSON, variables.css custom properties, and Tailwind v4 theme.css.
  • Asset capture for faithful reuse: downloads fonts, icons, images, logos, and per-page raw dumps for traceability and auditability.

Quick Start

Ask your agent to run the skill with a URL to extract, for example: /design-system-extract https://pirsch.io

Frequently Asked Questions about design-system-extract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract design tokens from a public website?

You can extract design tokens from a public website by crawling multiple internal pages and aggregating computed visual styles. This process captures colors, typography scale, spacing, radii, and component patterns from the site's rendered output into a reusable bundle.

Can I generate a Tailwind v4 theme from an existing website's CSS variables?

Yes, you can generate a Tailwind v4 theme from a website's CSS variables by scraping the rendered output. The extraction synthesizes computed styles into a Tailwind v4 theme.css file alongside DTCG-style tokens JSON and variables.css custom properties.

Does design system extraction work with Playwright for web scraping?

Design system extraction requires deterministic Playwright web scraping to crawl a site's homepage and discovered internal links. It captures rendered HTML, computed styles, screenshots, and referenced assets for faithful reuse and traceability.

What's the best way to capture a website's fonts and assets for a design system?

The best way to capture a website's fonts and assets is through multi-page crawling that downloads fonts, icons, images, logos, and per-page raw dumps. This ensures faithful reuse and provides full traceability to the raw extraction data.

Does web scraping for design tokens require manual style inference?

Web scraping for design tokens eliminates manual style inference by automatically extracting computed visual styles across multiple pages. It aggregates colors, typography, spacing, radii, and component patterns directly from the site's rendered output.

What format does the extracted design system bundle use for output?

The extracted design system bundle outputs to a DESIGNS/<hostname>/ directory containing four files: a narrative reference, DTCG-style tokens JSON, variables.css custom properties, and Tailwind v4 theme.css for immediate integration.