design-taste

Provides design direction, judgment heuristics, and anti-AI-slop review for crypto UI work.

Updated Mar 12, 2026
One-click install
npx skills add https://github.com/RavitejaKarra24/dotfiles --skill design-taste-ravitejakarra24
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-taste
Source: https://github.com/RavitejaKarra24/dotfiles/tree/main/agents/.agents/skills/design-taste
Command: npx skills add https://github.com/RavitejaKarra24/dotfiles --skill design-taste-ravitejakarra24

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated interfaces often look generic and interchangeable. This Skill supplies the design judgment layer between brand selection and implementation, helping you choose a concrete aesthetic direction, make layout and density decisions, and detect generic AI-slop patterns before shipping. ## Core Features & Use Cases - Design Briefs: Produces a compact design brief covering direction archetype, density, surface model, typography mood, motion, and explicit do/don't lists before any UI is built. - Anti-Slop Review: Audits existing UI against a two-tier detection checklist (strong and weak signals) to identify generic AI output, with enforceable code-scannable rules like banning pure black and transition-all. - Theme & Archetype References: Ships six style references (Warm Monochrome, Stark Minimal, Gradient Trust, Workstation Dense, Soft Consumer, Gallery Editorial), page archetypes, decision trees, and pitch deck visual directions. - Use Case: When a user says "this dashboard looks AI-generated", the Skill runs review mode, flags slop signals, and outputs a redesign brief such as a Workstation Dense direction with compact density and monospace numerals. ## Quick Start Ask the assistant to review your UI for generic AI patterns and propose a stronger design direction with a design brief.

Frequently Asked Questions about design-taste

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

FAQPage Schema
How do I make my UI look less AI-generated?

Run the anti-slop review mode, which scans for strong signals like uniform border-radius, centered symmetric layouts, default gradients, and card nesting. If three or more strong signals match, the Skill recommends committing to one of 11 aesthetic directions and outputs a redesign brief.

What design direction should I choose for a crypto dashboard?

The Skill maps product types to style references: trading interfaces start with Workstation Dense, fintech products with Gradient Trust, and SaaS tools with Warm Monochrome. It then produces a design brief specifying density, surface model, typography mood, and motion.

When should I use design-taste versus frontend-design-guidelines?

Use design-taste for direction questions like what a page should look like, density choices, and slop review. Use frontend-design-guidelines for mechanical correctness such as accessibility, forms, and states; it delegates direction questions to this Skill.

Does this Skill override my existing brand palette or typography?

No. It reads brand.md first and respects palette and typeface choices made by brand-design. Anti-slop rules target thoughtless defaults, not deliberate brand decisions documented in brand tokens.

What are the enforceable rules this Skill checks in code?

Three code-scannable rules: no pure black #000 in product UI unless documented as a brand token, no transition-all or transition: all, and no arbitrary Tailwind spacing values like p-[ or gap-[ patterns. Other heuristics require manual visual judgment.