skills.rest
kartikkabadi
kartikkabadiCOMMUNITY
Explore all skills by this author →
𝕏@1kartikkabadi1GitHub

stitch-design-taste

Generate a semantic DESIGN.md with atmosphere, palette, typography, layout, and motion constraints.

Updated Jul 17, 2026
Search Tags:#typography#design system#ui design#motion#responsive layout#stitch
One-click install
npx skills add https://github.com/kartikkabadi/skills --skill stitch-design-taste-kartikkabadi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: stitch-design-taste
Source: https://github.com/kartikkabadi/skills/tree/main/stitch-design-taste
Command: npx skills add https://github.com/kartikkabadi/skills --skill stitch-design-taste-kartikkabadi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you translate a project’s visual direction into a rigorous DESIGN.md that Google Stitch can use to generate polished, non-generic interfaces with consistent taste, structure, and motion.

Core Features & Use Cases

  • Atmosphere definition: Establishes the mood, density, variance, and motion style so the generated UI feels intentional instead of default.
  • Color and typography rules: Defines strict palette, font, hierarchy, and anti-pattern constraints to keep outputs premium and brand-consistent.
  • Layout and motion guidance: Encodes grid behavior, responsive collapse rules, component styling, and micro-interactions for high-quality screen generation.
  • Use case: Create a DESIGN.md for a dashboard, marketing site, or product surface that Stitch can follow to generate a cohesive visual system.

Quick Start

Ask the skill to generate a DESIGN.md for your project style, target screens, and preferred visual mood.

Frequently Asked Questions about stitch-design-taste

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

FAQPage Schema
How do I create a design system specification for AI-driven UI generation?▼

To create a design system specification for AI-driven UI generation, define a semantic DESIGN.md covering atmosphere, palette, typography, layout, and motion constraints to enforce non-generic responsive outputs.

What is a Stitch-ready design system and when do I need one?▼

A Stitch-ready design system is a rigorous DESIGN.md file encoding visual direction, density, and motion rules. You need one to generate cohesive, premium interfaces instead of default generic UI.

How do I set up typography and color rules for a design system?▼

Set up typography and color rules by defining strict palette, font hierarchy, and explicit anti-pattern constraints within your design specification to keep generated screens brand-consistent and premium.

Can I generate a responsive layout design system for a dashboard?▼

Yes, you can generate a responsive layout design system for a dashboard by encoding grid behavior, responsive collapse rules, and component styling to guide high-quality screen generation.

Does this approach prevent generic UI outputs from AI screen generation?▼

Yes, preventing generic UI outputs requires applying strict design-system constraints with explicit values and guardrails, ensuring the generated interface maintains intentional taste and structure.

What's the best way to encode motion and micro-interactions in a design specification?▼

The best way to encode motion in a design specification is to define mood, motion style, and micro-interaction rules alongside layout guidance, ensuring generated UI feels intentional rather than default.

Related Skills

Explore More Design & Creative Skills→
hi-yuki-922hi-yuki-922
1

i-teach-impeccable

Gather design context from codebase and user input into CLAUDE.md.

Community
Intermediate
efekrbasefekrbas
41

stitch-design-taste

Generate Stitch-compatible DESIGN.md files from frontend product briefs.

Community
Advanced
XCmiaowXCmiaow

stitch-design-taste

Convert a chosen UI taste profile into a DESIGN.md specification for Google Stitch.

Community
Advanced
fysoul17fysoul17
4

devlyn:design-system

Extract concrete design tokens from HTML style previews into framework-agnostic design references.

Community
Intermediate