design-intelligence

Synthesize project memory and design tokens into evidence-grounded UX recommendations.

Updated Apr 27, 2026
One-click install
npx skills add https://github.com/vTRKA/supervibe --skill design-intelligence-vtrka
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-intelligence
Source: https://github.com/vTRKA/supervibe/tree/main/skills/design-intelligence
Command: npx skills add https://github.com/vTRKA/supervibe --skill design-intelligence-vtrka

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Prevents generic, unverified design recommendations by grounding style and UX decisions in project memory, code facts, approved design-system tokens, and design-intelligence evidence packages.

Core Features & Use Cases

  • Evidence-backed design direction: synthesizes style, UX, brand/collateral, charts, and stack UI guidance with explicit citations and conflict handling.
  • Anti-slop, fail-closed gating: blocks when required proof, approvals, validator output, or runtime receipt is missing, and routes to owning specialist routes for advanced visual work (motion/media, charts, browser-runtime).
  • Design intelligence evidence packets: emits structured “Design Intelligence Evidence” including advanced visual overlays and required gate ids for safe handoff.

Quick Start

Ask the AI agent to run /supervibe-design with your design intent and the source artifact paths you want grounded in evidence.

Frequently Asked Questions about design-intelligence

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

FAQPage Schema
How do I ground UX decisions in project evidence instead of generic recommendations?

Evidence-based UX recommendations are generated by synthesizing project memory, codebase facts, and approved design-system tokens. The system blocks unverified design outputs when artifact proof, approvals, or runtime receipts are missing.

How do I enforce an anti-AI-slop contract for design system tokens?

An anti-AI-slop contract enforces fail-closed gating by requiring exact gate taxonomy ids before producing design outputs. It blocks decisions when required proof, validator output, or runtime receipts are missing.

What's the best way to verify brand collateral decisions against approved design tokens?

Brand collateral decisions are verified by running a strict preflight order: memory lookup, then code search, then internal design-intelligence pack lookup. This produces decision outputs with explicit citations and conflict handling.

Can I use design intelligence packs for advanced visual surfaces like motion and media?

Design intelligence packs apply to advanced visual surfaces including motion and media. The system routes these specialized tasks to owning specialist routes and emits structured evidence packets with required gate ids for safe handoff.

Why are my evidence-grounded design recommendations returning partial or blocked results?

Recommendations return blocked or partial outputs when artifact proof, approvals, or runtime receipts are missing. The fail-closed gating mechanism prevents generic design outputs by requiring complete evidence before returning decisions.

Do I need project memory and codebase facts to generate UI token guidance?

Project memory and codebase facts are required inputs for generating UI token guidance. The strict preflight order processes memory first, then code search, then internal lookup to synthesize verified stack UI guidance.