stitch-design-taste

Generate DESIGN.md files encoding design rules for Google Stitch.

53|25|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/naveedharri/benai-skills --skill stitch-design-taste-naveedharri
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stitch-design-taste
Source: https://github.com/naveedharri/benai-skills/tree/main/plugins/design/skills/stitch
Command: npx skills add https://github.com/naveedharri/benai-skills --skill stitch-design-taste-naveedharri

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Design teams struggle to maintain consistent, premium UI prompts across Stitch workflows, leading to inconsistent outputs and time wasted on manual spec writing.

Core Features & Use Cases

  • Generates DESIGN.md files tailored for Google Stitch to drive screen generation with a single source of truth
  • Encodes atmosphere, color, typography, layout, and motion rules into a repeatable design prompt
  • Suitable for design system teams, product squads, and AI-assisted UI generation pipelines

Quick Start

Run the design taste workflow to generate a DESIGN.md for your Stitch project by providing a target screen prompt and atmosphere.

Frequently Asked Questions about stitch-design-taste

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

FAQPage Schema
How do I generate a DESIGN.md file for Google Stitch?

To generate a DESIGN.md file for Google Stitch, provide a high-level design goal and target screen prompt. The workflow applies atmosphere, color, typography, layout, and motion rules to output a structured design prompt.

What is a Stitch design prompt and when do I need one?

A Stitch design prompt is a structured DESIGN.md file encoding exact token rules, color palettes, and typography constraints. You need one to maintain consistent, premium UI generation across AI-assisted screen generation pipelines.

How do I enforce UI specs and prevent anti-patterns in design system workflows?

You enforce UI specs and prevent anti-patterns by encoding anti-pattern bans and exact token rules into a structured DESIGN.md. This creates a single source of truth for design agents to follow during screen generation.

Can I use a DESIGN.md to standardize color and typography tokens for AI UI generation?

Yes, you can standardize color and typography tokens by generating a DESIGN.md that encodes exact color palettes and typography constraints. This ensures consistent, reusable prompts for Stitch-driven UI generation pipelines.

Does this approach work for design system teams needing reusable screen generation prompts?

Yes, this approach works for design system teams by producing repeatable design prompts. It encodes atmosphere, layout, and motion rules into a single source of truth, eliminating time wasted on manual spec writing.