brand-design

Generate, preview, and apply brand color palettes, typography, and gradients to shadcn frontend projects.

Updated Jul 10, 2026
One-click install
npx skills add https://github.com/Kaleb-Rupe/aurora --skill brand-design-kaleb-rupe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brand-design
Source: https://github.com/Kaleb-Rupe/aurora/tree/main/claude/skills/brand-design
Command: npx skills add https://github.com/Kaleb-Rupe/aurora --skill brand-design-kaleb-rupe

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Newly scaffolded frontend projects ship with the default shadcn neutral theme and no visual identity, and picking a cohesive palette, font pairing, and gradients by hand is slow and error-prone. This Skill runs a structured brand pass that interviews the user, previews six candidate palettes in the browser, and writes the result directly into the project's CSS. ## Core Features & Use Cases - Interactive palette picker: Generates 6 candidate palettes (3 curated from 30 recipes, 3 algorithmic in OKLCH) and renders them in a self-contained HTML preview with a contextual mini-UI matching the product category (wallet, swap, staking, NFT, dashboard, social). - WCAG AA enforcement: Every palette is contrast-verified against the full shadcn token set in both light and dark mode, with automatic lightness adjustment or regeneration on failure. - Full application pipeline: Writes shadcn CSS variables to app/globals.css (with backup), wires Google Fonts via next/font, optionally derives two brand gradients, and produces a brand.md artifact for future sessions. - Use Case: After scaffolding a Next.js + Tailwind + shadcn app, ask for a brand identity; the Skill interviews you about category and mood, opens a browser preview of 6 palettes rendered on your actual product UI, then applies your pick with typography and a brand.md reference file. ## Quick Start Ask the assistant to run the brand-design skill to pick brand colors and a theme for this project.

Frequently Asked Questions about brand-design

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

FAQPage Schema
How do I generate a brand color palette for a shadcn project?

Run the brand-design skill, which interviews you about product category and mood, then generates 6 candidate palettes rendered in a browser preview. Pick one by number and it writes the full shadcn token set to app/globals.css in light and dark mode.

How to preview color palettes before applying them to a Next.js app?

The skill writes a self-contained HTML file to .brand-preview/index.html and opens it in your browser. It renders the same contextual mini-UI (wallet, swap, dashboard, etc.) six times with different palette tokens, plus a light/dark toggle.

Does the palette generation check WCAG contrast accessibility?

Yes, every palette is verified against WCAG AA before being shown: 4.5:1 for body text pairs and 3.0:1 for large text and icons. Failing pairs are auto-adjusted in lightness up to three times, or the palette is dropped and regenerated.

Can I use this with a project that is not Next.js and Tailwind?

Yes, but the apply step adapts: instead of writing shadcn CSS variables, it writes the palette to brand.md only and explains how to map the tokens into your stack manually. The interview, preview, and contrast checks still work.

What happens to my existing globals.css when a new theme is applied?

The file is always backed up to app/globals.css.bak before any write. Only the :root and .dark token blocks are replaced; Tailwind imports, custom layers, and other content are preserved, with rollback from the backup if verification fails.

How does typography pairing work with the chosen palette?

After palette selection, six Google Fonts pairs matched to the palette's mood are rendered in a second HTML preview on your actual brand colors. Your pick is wired via next/font/google in the root layout, defaulting to Inter plus JetBrains Mono if skipped.