About design-system-extractor
A skill for extracting a reusable design system from scattered UI, intended for teams standardizing an existing product interface. It inventories tokens, components, states, and patterns, then produces a system map with gaps, duplicates, and validation checks.
Key features
- 01Inventory UI tokens and components
- 02Capture states and layout patterns
- 03Normalize names and distinguish primitives
- 04Identify gaps and duplicated elements
- 05Create a migration and validation plan
Use cases
- 01Document an existing product's design system
- 02Find duplicated UI components across surfaces
- 03Plan migration toward canonical component names
- 04Check accessibility and responsive design coverage
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ sylphxai/ skills --skill design-system-extractorAfter installing, type/design-system-extractorin Claude Code to confirm it loaded.
Bundled filesagents/openai.yamlreferences/design-system-extraction-systems.md
Details
From the project READMEDesign System Extractor
Use this skill to turn scattered UI into a coherent design system an agent can reuse.
Workflow
- Identify source surfaces, product type, platforms, brand constraints, and target consumers of the system.
- Read
references/design-system-extraction-systems.md. - Inventory tokens, components, states, layout patterns, content patterns, motion, accessibility, and exceptions.
- Normalize into canonical names and separate primitives from product-specific components.
- Produce a design-system map with gaps, duplicates, migration plan, and validation checks.
When not to use
- Do not use for generic advice the base model already handles without this skill's specific artifact contract.
Guardrails
- Do not invent a new system when the task is to extract the existing one.
- Do not collapse meaningful product variants into one generic component.
- Do not ignore accessibility states, error states, density, responsive behavior, or motion.
Output format
Source surfaces:
System shape:
Tokens:
- color / type / spacing / radius / elevation / motion
Components:
- <component> -> variants, states, usage, gaps
Inconsistencies:
- <issue> -> recommended canonical pattern
Migration plan:
- quick wins / systematic fixes / open questions
Comments
Markdown supported