design-md-writer

Render a canonical DESIGN.md from extracted brand tokens using a Jinja2 template.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/imehr/design-extractor --skill design-md-writer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-md-writer
Source: https://github.com/imehr/design-extractor/tree/main/skills/design-md-writer
Command: npx skills add https://github.com/imehr/design-extractor --skill design-md-writer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Render a consistent, auditable brand DESIGN.md from scattered extraction artefacts so designers and engineers can review tokens, patterns, assets, and validation evidence without manual assembly.

Core Features & Use Cases

  • Template-driven rendering: Produces a canonical markdown using a Jinja2 template to ensure consistency across brands.
  • Evidence-rich documentation: Inlines confidence badges, validation scores, and provenance links alongside token and pattern tables for auditing.
  • Re-render and install flow: Builds a YAML-frontmatter-backed DESIGN.md that can double as an installable per-brand SKILL.md and be re-generated when tokens change.
  • Use Case: Post-extraction documentarian step to convert the cache into a single, reviewable DESIGN.md for handoff or inclusion in a design system library.

Quick Start

Render the DESIGN.md for the brand with slug <slug> from the extraction cache located at ~/.claude/design-library/cache/<slug>/ using the design-md-writer template.

Frequently Asked Questions about design-md-writer

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

FAQPage Schema
How do I generate a DESIGN.md from design system tokens and patterns?

To generate a DESIGN.md from design system tokens and patterns, use a template-driven renderer to apply a Jinja2 template against extracted brand caches, producing a canonical markdown document with confidence badges and evidence tables.

What is the best way to document brand tokens and validation evidence for developer handoff?

Documenting brand tokens and validation evidence for developer handoff is best handled by rendering a YAML-frontmatter-backed markdown file that inlines validation scores and provenance links alongside token tables for auditable consistency.

How do I render a canonical brand document using Jinja2 templates?

Render a canonical brand document using Jinja2 templates by supplying extraction artifacts like tokens-output.json, patterns.json, and voice-analysis.json to a DESIGN.md.jinja template to produce structured markdown.

Do I need extraction artifacts to generate a template-driven brand document?

Yes, you need extraction artifacts to generate a template-driven brand document, specifically requiring tokens-output.json, patterns.json, voice-analysis.json, report.json, and metadata.json from the brand cache.

Can I use the rendered DESIGN.md as an installable per-brand skill?

Yes, you can use the rendered DESIGN.md as an installable per-brand skill because the output is a YAML-frontmatter-backed markdown file that doubles as an installable SKILL.md for design system libraries.

When should I re-render a template-driven brand document?

You should re-render a template-driven brand document when extracted brand tokens change, ensuring the canonical DESIGN.md reflects updated validation evidence and pattern data without manual assembly.