neo-brutalist-ui-system

Apply a reusable neo-brutalist design system to existing or new frontend projects (landing, marketing, portfolio, dashboard, app surfaces) using strong typography, tokenized theming, and variant-driven component primitives. Preserve existing information architecture unless restructuring is explicitly requested.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "neo-brutalist-ui-system" with this command: npx skills add devfirexyz/ui-skills/devfirexyz-ui-skills-neo-brutalist-ui-system

Neo-Brutalist UI Design System Skill

Use this skill when users ask for bold, high-contrast UI refactors, reusable design systems, or full-page styling upgrades.

This skill is design-system-first, not template-first.

  • Default behavior: preserve existing layout and content architecture.
  • Rebuild layouts only when explicitly requested.

Core Principles

  • Typography drives hierarchy more than decoration.
  • Strong structure: 2px borders, explicit separators, deliberate spacing.
  • Hard-edged surfaces: square corners by default on primary UI.
  • Direct interactions: translate + shadow reduction for active states.
  • Tokenize everything important (background/surface/text/border/accent/shadow).
  • Reuse existing app theme when present; generate missing tokens only.

Design DNA Extraction (Required First Step)

Before editing, extract project design DNA from codebase:

  1. Global styling source
  • find global CSS/theme entry (app.css, globals.css, theme.css).
  • collect color variables, font declarations, theme toggling strategy.
  1. Typography roles
  • identify display/headline font, body font, technical/mono font, optional accent font.
  • map where each role is used (hero, nav, cards, metadata, captions).
  1. Primitive inventory
  • locate core UI primitives (button, card, input, textarea, dialog, dropdown, tabs, tooltip, progress, avatar, separators).
  • detect variant API style (cva, prop variants, class maps).
  1. Layout rhythm
  • identify section cadence (bordered slices, alternating surfaces, grid separators, CTA blocks).
  • identify motion style (hover transitions, reveal animations, duration patterns).
  1. Constraints
  • preserve information architecture unless user asks for restructure.
  • preserve brand tone and color intent unless user asks to rebrand.

Scope Modes

  • component-only: update tokens and primitive components only.
  • layout-preserving (default): restyle existing pages, keep same sections/content flow.
  • layout-rebuild: redesign section architecture (explicit user request required).

Reference Pattern Library (Derived from Lawn, Applied Generically)

Typography System

  • Display/headlines:
    • very high weight (800-900)
    • tight tracking
    • uppercase for key headings/CTAs where tone supports it
  • Body/support copy:
    • cleaner medium/regular weight
    • reduced contrast tone for secondary text
  • Technical/meta text:
    • monospaced role for timestamps, metrics, status labels
  • Optional accent typography:
    • serif or stylistic contrast font for specific emphasis only (not everywhere)

Token Model

Prefer this token family so output works across page types:

  • --background, --background-alt
  • --surface, --surface-alt, --surface-strong, --surface-muted
  • --foreground, --foreground-muted, --foreground-subtle, --foreground-inverse
  • --border, --border-subtle
  • --accent, --accent-hover, --accent-light
  • --shadow-color, --shadow-accent
  • semantic tokens: --destructive, --success, --warning

Theme Behavior

  • Support both light and dark token sets when the project already supports them.
  • Use data-theme switching for deterministic theme control where possible.
  • If project already persists theme state (local storage or framework state), preserve that flow.
  • Do not force dark mode redesigns on light-first products unless requested.

Primitive Patterns

  • Buttons:
    • bold label, often uppercase
    • border-2
    • hard shadow
    • hover/active translate by 1-2px
  • Cards:
    • strong border + structured internal header/content/footer spacing
    • title emphasis via weight/tracking, not ornament
  • Inputs/Textareas:
    • hard-edged, border-forward, clear focus border/shadow
    • mono/meta-friendly when context is technical
  • Overlay components (dialog/dropdown/tooltip):
    • maintain Radix accessibility behavior
    • apply same border/shadow vocabulary as core surfaces
  • Data components (tabs/progress/badges):
    • high-legibility state contrast
    • avoid soft/ambiguous state styling
  • Exception rule:
    • circular avatars are allowed when functionally expected.

Layout Patterns

Apply across landing/marketing/portfolio/dashboard:

  • Section rhythm via border dividers and controlled spacing blocks.
  • Grid separators (divide-x/divide-y or explicit borders) for feature/data matrices.
  • Contrast bands (light vs dark or muted vs strong surfaces) to define narrative steps.
  • CTA zones with the strongest type/contrast, not necessarily the loudest color.
  • Dashboard/application screens prioritize utility clarity over marketing theatrics.

Motion Patterns

  • Use fast, explicit transitions (120-200ms) for hover/active state changes.
  • Use 1-2 meaningful entrance/reveal patterns max.
  • Avoid decorative animation noise that does not improve hierarchy or feedback.

Workflow

1. Choose Intent + Scope

  • Define one-line aesthetic intent (example: "industrial editorial", "clean blocky utility").
  • Choose scope mode (default layout-preserving).

2. Establish/Map Tokens

  • If project already has tokens: map to the token model above and keep naming conventions.
  • If project lacks tokens: generate with script, then integrate.

Script: python3 scripts/generate_blocky_palette.py --base "<hex>" --accent "<hex>" --format both

3. Refactor Primitives First

  • Update primitives before page-level classes.
  • Prefer variant APIs over one-off per-page styles.
  • Keep class composition consistent (cva + shared util) when stack supports it.

4. Apply to Layouts

  • layout-preserving: restyle existing sections without changing narrative/order.
  • layout-rebuild: propose new architecture first, then implement after alignment.
  • For dashboards/apps: emphasize information density control and scanability.
  • For landing/marketing/portfolio: emphasize hierarchy and narrative rhythm.

5. QA Pass

  • Responsiveness: no broken borders/overlaps at mobile/tablet/desktop.
  • Contrast: readable foreground/background across light and dark contexts.
  • Interaction: focus-visible states are clear; hover/active states remain intentional.
  • Consistency: primitives and page sections use same token/shadow/border language.

Output Contract

Return:

  • Design DNA summary (fonts, token map, primitive inventory, layout rhythm found).
  • Changes implemented (primitives and/or layouts) with preserved vs rebuilt statement.
  • Token block updates (or mapping strategy when existing theme is reused).
  • Short rationale on hierarchy, readability, and interaction clarity.

Script Usage After Install

When installed, the script is bundled in this skill:

  • Agent run pattern:
    • python3 scripts/generate_blocky_palette.py --base "<hex>" --accent "<hex>" --format both
  • Manual run pattern:
    • cd <installed-skill-path>/neo-brutalist-ui-system
    • python3 scripts/generate_blocky_palette.py --base "#f2f1eb" --accent "#2f5f3a" --format both

Use script output as source-of-truth tokens, then map into project naming if needed.

Example Prompts

  • "Use $neo-brutalist-ui-system in layout-preserving mode; keep my current sections and apply a full design-system refactor."
  • "Use $neo-brutalist-ui-system in component-only mode and standardize button/card/input/dialog/dropdown/tabs to one tokenized style."
  • "Use $neo-brutalist-ui-system for my dashboard; keep the current IA, improve scanability, typography hierarchy, and interaction clarity."

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

dexiejs

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

inference-agent-ui

No summary provided by upstream source.

Repository SourceNeeds Review
General

baseline-ui

No summary provided by upstream source.

Repository SourceNeeds Review
3.7K-ibelick
General

fixing-motion-performance

No summary provided by upstream source.

Repository SourceNeeds Review
3.6K-ibelick