What is the design intelligence skill?
RYUX, in the design-intelligence repo, is a design skill that makes an AI agent ask about the product first, keep unknown prices and policies off the screen, design loading, error and offline states, and explain its choices. A scan command also checks code and rendered pages without an LLM.
RYUX is a design skill for AI agents published in the design-intelligence repository. Its pitch is to understand before designing: the agent asks about the product, keeps what it does not know out of the screens, designs for when things go wrong and explains each decision.
It ships as one skill folder, ryux, with a router SKILL.md, five capability files and knowledge modules. One workflow sits behind every step: understand, frame, explore alternatives, evaluate, build and critique, then verify.
What the README reports from its benchmarks
- In an order-tracking brief, agents without the skill showed 28 invented business rules as fact across three runs, against 1 with RYUX, counted blind.
- Offline, loading and error states appeared in every RYUX run and in none of the no-skill runs.
- A senior designer rated one checkout blind at 4.33 with RYUX against 2.83 without.
- One illustration benchmark failed at first, then passed a rerun after a fix.
- A RYUX run costs about 1.5 to 7 times a run without a skill.
Who the skill is for
Best for
- Prototypes where invented fees, promos and policies would mislead stakeholders.
- Teams that want critique with severity and evidence, like critique, plus a QA pass at several widths.
- Developers who want a no-LLM scan in CI for alt text, contrast and sideways scroll.
Not for
- Quick, cheap drafts. It asks first and costs more per run.
- Full reference-data features. The README says reference screens from real products are private, optional and in early access.
- Guaranteed rating gains; the benchmarks are two to three runs per side and describe those runs, not rates.
Use it with UiChemy
Run the Analyze or Critique step on the WordPress page you ship, and use the scan on a rendered URL. Build the page itself with UiChemy Composer or AI to WordPress. Click Get Started on uichemy.com.
How to install Design Intelligence Skill (RYUX): Ask Before You Design
npx @ryuxdsgn/design-intelligencePlaywright is needed in your project for URL scans. Codex runs project hooks only after you trust the project.Run the installer
Run the command in your project. The README says it works with Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, Cline, Amp, Kimi Code and Antigravity.
Optional setup
Run the setup subcommand to write DESIGN.md and your project's direction, or add the hook flag to install for edit-time findings.
Plugin route
In Claude Code run /plugin marketplace add ryuxdsgn/design-intelligence, then /plugin install ryux@design-intelligence. The guide covers Codex, Copilot, VS Code and Gemini CLI plugins.
Use it
Say what you are doing, such as "Critique our pricing page", and RYUX picks the step.
Check that it loaded
Ask for a checkout screen from a one-line brief. The agent should ask questions first, then build with unknown prices labelled and five states covered.
Key features
- Asks before it designsThe agent asks about direction, pricing, payment and what a screen may promise before drawing it.
- Unknowns stay unknownUndecided prices, fees and policies are labelled or left out instead of invented, a hard gate in the rules.
- Failure states by defaultLoading, offline, error and empty states are designed, not left as an afterthought.
- Five task modesAnalyze, Design, Build, Critique and QA, picked from what you say in plain words.
- Rule levels and hard gatesRules are Required, Preferred or Contextual, with gates that are never waived, such as no invented numbers and no missing critical states.
- Scan commandChecks source with no LLM or API key, and a URL at 320, 390 and 1280 px for sideways scroll, contrast, target size and colliding text.
- HooksThe install can add a hook so the agent gets findings for each file it edits, supported for four named agents in the README.
- Figma-aware workReads a file's own libraries as the design system and writes DESIGN.md from a brief or a Figma file.
- Report of what was verifiedEvery task ends with what was verified and what was not.
Frequently asked questions
What is RYUX design intelligence?
It is a free MIT skill for AI agents, in the design-intelligence repo, that asks about the product before it designs, avoids invented rules, designs failure states and explains each decision. It also includes a scan command for code and rendered pages.
How do I install it?
Run npx @ryuxdsgn/design-intelligence in your project. The README also lists a Claude Code plugin route, plugins for Codex, GitHub Copilot, VS Code and Gemini CLI described in its guide, and global installs and updates.
Which agents does it work with?
The README names Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, Cline, Amp, Kimi Code and Antigravity. Edit-time hooks are described only for Claude Code, Codex, Cursor and OpenCode, with different mechanics for each.
What does the scan command check?
Without an LLM or API key it checks source for issues like images with no alt text, unnamed icon buttons and urgency copy. Given a URL it renders at 320, 390 and 1280 px and measures sideways scroll, contrast, target size and colliding text.
Does it really reduce made-up product rules?
In the README's order-tracking benchmark agents without the skill showed 28 invented policies as fact across three runs, and with RYUX it was 1, counted blind. The README says benchmark numbers describe those runs, not rates.
Does RYUX cost more to run?
Yes. The README says a run costs about 1.5 to 7 times a run without a skill in its benchmarks because it asks first, checks states and writes its rationale.
What are related skills?
critique gives heuristic feedback on a screen, ux-heuristics covers Nielsen and Krug ratings with severity, and a11y-ally runs axe-core, pa11y and Lighthouse audits. design-md explains the DESIGN.md format RYUX can write.


