brainstorming
MUST USE when the user wants new features, behavior changes, refactoring with new capabilities, or architecture decisions and no…
MUST USE for any frontend, UI, or web interface implementation. Enforces production-grade visual quality, accessibility, responsive design, and professional polish. Triggers on: "build a UI", "frontend", "website", "landing page", "dashboard", "make it look professional", any
$ npx -y skills add REPOZY/superpowers-optimized --skill frontend-design --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/frontend-designContext preview
The summary Claude sees to decide when to auto-load this skill.
MUST USE for any frontend, UI, or web interface implementation. Enforces production-grade visual quality, accessibility, responsive design, and professional polish. Triggers on: "build a UI", "frontend", "website", "landing page", "dashboard", "make it look professional", any
name: frontend-design description: > MUST USE for any frontend, UI, or web interface implementation. Enforces production-grade visual quality, accessibility, responsive design, and professional polish. Triggers on: "build a UI", "frontend", "website", "landing page", "dashboard", "make it look professional", any React/Vue/Svelte/HTML component work. Routed by using-superpowers, or invoke directly via /frontend-design.
This skill transforms generic AI-generated UIs into production-grade, visually distinctive interfaces. It provides a design reasoning framework, industry-aware style selection, and concrete engineering standards.
1. Check: does the project already have a design system, component library, or style guide?
2. If YES (existing design system):
3. If NO (greenfield):
Before writing any frontend code, walk through this reasoning framework. The goal is to make deliberate design decisions — not default to generic blue-and-white SaaS templates.
Identify these before choosing any visual direction:
Use the Industry Reference Table (below) as a starting point, then refine based on project specifics.
Choose explicitly:
Every product type has styles that actively harm it. Name them explicitly before implementation:
Before writing code, output a brief design system summary:
**Design System**: [Product] — [Style] **Colors**: [Primary] / [Accent] / [Neutral] / [Semantic] **Typography**: [Heading font] + [Body font] — [Scale] **Effects**: [Key interactions] **Avoid**: [Named anti-patterns]
This takes 30 seconds and prevents hours of rework from misaligned visual direction.
---
Most AI-generated UIs fail here. They're technically correct but visually forgettable — the same blue-and-white SaaS template with rounded cards and a gradient hero. Professional UIs have identity.
The UI must include at least one of these distinctiveness signals:
Avoid these generic patterns:
**The test:** If the UI could belong to any startup with a search-and-replace on the logo and copy, it has failed. The design direction chosen in Step 2 should be visible in the final output.
---
25 styles with key characteristics. Use this to select and combine styles — most projects blend 1-2 primary styles.
| Style | Key Characteristics | Best For | Avoid When | |---|---|---|---| | **Minimalism / Swiss** | Monochromatic, geometric grid, essential elements only | Enterprise, dashboards, tools | Playful brands, children's products | | **Glassmorphism** | `backdrop-filter: blur(10-20px)`, translucent overlays, vibrant backgrounds | Modern SaaS, fintech dashboards | Low-performance targets, print | | **Brutalism / Neubrutalism** | Bold borders 2-4px, primary colors, thick offset shadows, no gradients | Creative agencies, portfolios, Gen-Z | Corporate, healthcare, finance | | **Neumorphism / Soft UI** | Dual shadows (light+dark), embossed/debossed, monochromatic, rounded | Wellness, smart home, controls | Data-dense, text-heavy, low contrast risk | | **Claymorphism** | Soft 3D, thick bord
The production-grade fork of obra/superpowers — same trusted workflow, dramatically leaner, safer, and more intelligent.
MUST USE when the user wants new features, behavior changes, refactoring with new capabilities, or architecture decisions and no…
Creates minimal, high-signal CLAUDE.md and AGENTS.md context files based on empirical best practices. Invoke on /init command,…
Use in long or noisy sessions to persist durable state across session boundaries via state.md. Also generates project-map.md when…
Use BEFORE brainstorming when facing a complex architectural, technology, or design decision where the options are not yet…
MUST USE when updating, migrating, or auditing project dependencies: upgrading packages, fixing security vulnerabilities (CVEs),…
Use when 2+ tasks are independent and can run concurrently without file or state conflicts or sequential dependencies. Triggers…