skills/design-intelligence/SKILL.md
Design and review web interfaces with trend-aware, performance-conscious, accessible rules. Triggers on design a page, build UI, review my site, audit this page, create a landing page, check my dashboard.
npx skillsauth add fatih-developer/fth-skills design-intelligenceInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
AI-powered web design skill that guides both generation and review of web interfaces. Works for any framework (HTML/CSS, React, Vue, Svelte, Next.js, Nuxt, etc.).
This skill operates in two modes. Identify the request type and apply the corresponding instructions.
Trigger: User wants to design, build, create, or scaffold a web page or UI.
Approach:
references/WEB_RULES.mdTrigger: User wants to review, audit, critique, or check an existing page or design.
Approach:
references/REVIEW_RUBRIC.md for detailed category guidanceUser request
│
├─ "tasarla / oluştur / build / create" → GENERATE Mode
│ └─ What page type? (landing, dashboard, e-commerce, blog, portfolio, pricing, about)
│
└─ "incele / audit / review / kontrol et / eleştir" → REVIEW Mode
└─ Is this a live URL or code/snippet?
├─ Live URL → navigate, snapshot, score
└─ Code/snippet → analyze directly
If not already clear, ask or infer:
Output:
## Page Strategy Summary
- Page type: [type]
- Page goal: [what the page must accomplish]
- Primary audience: [who uses this page]
- Primary message: [the one thing this page must communicate]
- Primary CTA: [the main action — can be "none" for purely informational pages]
List ordered sections for the page type. Example for landing page:
1. Hero — single message + primary CTA
2. Trust indicators — logos, stats, social proof
3. Feature bento grid — modular card cluster
4. Results or stats — measurable impact
5. Testimonial or proof — human evidence
6. FAQ — objection handling
7. Final CTA — last conversion push
Example for dashboard:
1. Navigation shell — sidebar or top bar
2. KPI cards — top-level metrics
3. Primary data view — main table/chart
4. Secondary panels — detail, filters, activity
5. Action bar — bulk actions, exports
Confirm each rule is satisfied. Reference references/WEB_RULES.md for the full rule text.
| Rule | Status | Notes | |------|--------|-------| | One Primary Message | ✅ | | | No Hero Slider | ✅ | | | Modular Layout | ✅ | | | Mobile-First | ⚠️ | Side nav collapses to hamburger on mobile | | Performance Targets | ✅ | LCP < 2.5s, INP < 200ms, CLS < 0.1 | | Accessibility Baseline | ✅ | Semantic HTML, focus states, contrast | | Functional Motion Only | ✅ | | | Forbidden Defaults | ✅ | No slider, no carousel, no decoration-only animation |
Only after the base solution is solid. Apply references/WEB_RULES.md Section "Advanced Effects":
Score each category 1-5 and record justification:
| # | Category | Score (1-5) | Justification | |---|----------|-------------|---------------| | 1 | Message Clarity | | | | 2 | Hero Focus | | | | 3 | CTA Clarity | | | | 4 | Modular Structure | | | | 5 | Visual Hierarchy | | | | 6 | Mobile Usability | | | | 7 | Touch Comfort | | | | 8 | Performance Safety | | | | 9 | Accessibility Quality | | | | 10 | Motion Usefulness | | | | 11 | Visual Consistency | | | | 12 | Advanced Effects Justification | | |
Reference references/REVIEW_RUBRIC.md for detailed scoring guidance per category.
Classify every issue into one of three tiers:
Critical — damages clarity, usability, accessibility, performance, or conversion path:
Important — weakens quality without breaking the page:
Nice to Improve — polish opportunities:
## Final Recommendation: [approve / revise / redesign]
[approve] — No critical issues, page is clear and usable
[revise] — No structural failure, but important issues weaken effectiveness
[redesign] — Structurally confused, hero strategy broken, mobile poor, systemic problems
Rules adapt based on page type. The default rules in references/WEB_RULES.md apply to all types. This matrix highlights per-type emphasis.
| Page Type | Primary Focus | Key Adaptation | |-----------|--------------|---------------| | Landing page | Conversion, single message | Hero is critical, CTA placement paramount | | Dashboard / Admin | Data density, task efficiency | High information density OK, hierarchy by importance | | E-commerce | Product discovery, conversion | Product hero, gallery, clear pricing, CTA priority | | Blog / Article | Readability, content first | Typography-first, minimal chrome, distraction-free | | Portfolio | Work showcase, personality | Visual-first, project cards, clean layout | | SaaS Pricing | Plan comparison, decision support | Table or grid layout, feature clarity, CTA per plan | | About / Contact | Trust, human connection | Team, story, clear contact channels | | Error / Empty states | Clarity, next action | Friendly message, clear CTA, recovery path |
Dashboard / Admin panel:
E-commerce product page:
Blog / Article:
Unless explicitly justified and explainable, these are never acceptable:
A forbidden default may be broken only if the agent:
If that explanation cannot be made convincingly, keep the default rule.
GENERATE workflow:
1. Identify page type → 2. Ask essential questions → 3. Page Strategy Summary
→ 4. Section Architecture → 5. UX Rule Check → 6. Optional Enhancements
REVIEW workflow:
1. Gather info (URL or code) → 2. Score 12 categories (1-5)
→ 3. Classify issues → 4. Final recommendation (approve/revise/redesign)
Core reading:
references/WEB_RULES.md — 27+ rules with per-type adaptationsreferences/REVIEW_RUBRIC.md — 12 scoring categories with detailed guidancereferences/EXAMPLES.md — concrete examples per page typetools
Create, optimize, critique, and programmatically structure prompts for AI systems. Use this skill whenever the user is designing or improving a static prompt, system prompt, coding prompt, agent prompt, workflow prompt, MCP-oriented prompt package, or an algorithmic prompt optimization pipeline. Also use it when the user asks to turn vague AI behavior into a precise instruction set, tool policy, agent spec, evaluation metric, or prompt architecture.
testing
Assumption-first architecture review skill to stress-test project plans and expose hidden risks.
testing
Enforce and manage DESIGN.md specifications, extract design systems from URLs, and combine design reasoning with token roles to prevent drift.
testing
Forces the agent to act with a Claude-like product mindset, prioritizing user journey, UX states, and visual quality before coding.