design

Lock user-approved aesthetic direction before generating frontend code or mockups.

5|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/barkleesanders/claude-code-starter --skill design-barkleesanders
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design
Source: https://github.com/barkleesanders/claude-code-starter/tree/main/skills/design
Command: npx skills add https://github.com/barkleesanders/claude-code-starter --skill design-barkleesanders

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill eliminates the risk of generic, low-quality AI-generated frontend work that lacks unique aesthetic direction, ensuring every interface has a deliberate, memorable design identity before any code or mockup is created.

Core Features & Use Cases

  • Two-Phase Direction-First Workflow: Locks a user-approved aesthetic direction (purpose, tone, constraints, differentiation) before routing work to either Stitch cloud generation or direct in-repo code writing.
  • Built-In Anti-Slop Guardrails: Enforces strict rules banning generic defaults like overused fonts, purple-on-white gradients, identical card grids, and other clichéd AI-generated patterns.
  • Preview-First Redesigns: For existing interfaces, generates a Stitch mockup for user approval before porting changes to production code, catching layout issues early.
  • Use Case: A frontend developer building a client-facing dashboard can use this Skill to pick a luxury refined direction, generate a Stitch mockup for stakeholder sign-off, then produce production React code that matches the approved aesthetic exactly.

Quick Start

Use the design skill to build an editorial magazine-style blog homepage in my existing Astro repo.

Frequently Asked Questions about design

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

FAQPage Schema
How do I prevent generic AI-generated frontend design in my UI mockups?

To prevent generic AI-generated frontend design, you must lock a user-approved aesthetic direction before generating any UI mockups or code. This approach enforces strict anti-slop guardrails that ban overused fonts and identical card grids, ensuring a deliberate design identity.

What is the process for redesigning an existing interface without breaking production code?

Redesigning an existing interface safely requires a preview-first workflow. You generate a UI mockup for user approval first, which catches layout issues early before porting the approved aesthetic changes directly into your production frontend code.

Does frontend design generation support direct in-repo code writing or only cloud mockups?

Frontend design generation supports both direct in-repo code writing and cloud mockups. It routes work to Stitch cloud generation or writes code directly in your repository based on your intent, delivering production-grade, accessible code that passes deterministic validation.

How do I create a design system that avoids clichéd AI patterns like purple-on-white gradients?

Creating a design system that avoids clichéd AI patterns requires enforcing strict anti-slop rules during development. The system bans generic defaults like purple-on-white gradients, mandates OKLCH color values, and validates against anti-patterns before completion.

Can I use UI mockup generation for an existing Astro repo?

Yes, you can use UI mockup generation for an existing Astro repo. The workflow applies a locked aesthetic direction to your project, routing the frontend design work to either cloud generation or direct in-repo code writing to deliver production-grade code.