К основному содержанию
плагин

ux-ui-agent-skills — плагин для Claude Code

Плагин ux-ui-agent-skills для Claude Code добавляет токены дизайна, спецификации и 52 гейта для проверки вёрстки по стандартам WCAG 2.2.

MIT ★ 3 216 обновлён 2026-10-20

Установить в Claude Code · Через маркетплейс плагинов Все способы и платформы ↓
/plugin marketplace add plugin87/ux-ui-agent-skills
/plugin install ux-ui-agent-skills@ux-ui-agent-skills
Открыть на GitHub

Что делает ux-ui-agent-skills

Плагин добавляет в Claude Code инструкции архитектора интерфейсов, 26 готовых скиллов и спецификации для 52 компонентов. База включает токены DTCG и параметры 2,144 дизайн-систем, большинство из которых замерены прямо с работающих сайтов. Код генерируется под разные стеки через 16 адаптеров фреймворков.

Главное отличие от обычных промптов — 52 объективных гейта качества. 34 из них открывают браузер для проверки контраста WCAG, доступности axe, фокуса с клавиатуры, размера кликабельных зон и отсутствия горизонтальной прокрутки на ширине 280 пикселей. Ошибки валят сборку, поэтому агент не выдаст небрежный интерфейс за готовый результат.

Пакет не требует сборщиков или сторонних библиотек при установке. Это слой правил и знаний, работающий напрямую в контексте модели.

Какие задачи решает

  • убрать типичные ошибки агентов с нечитаемыми градиентами и одинаковыми карточками
  • проверить доступность разметки по критериям WCAG 2.2 AA и AAA
  • генерировать компоненты по спецификациям DTCG без сторонних рантайм-зависимостей
  • остановить пайплайн при непройденных тестах вёрстки на мобильной ширине 280px

Примеры запросов агенту

  • Собери экран аналитики через команду /data-dashboard
  • Создай форму настроек по дизайн-токенам DTCG с проверкой доступности
  • Проверь контрастность состояний кнопки через встроенные гейты качества

Как установить ux-ui-agent-skills

Через маркетплейс плагинов

/plugin marketplace add plugin87/ux-ui-agent-skills
/plugin install ux-ui-agent-skills@ux-ui-agent-skills

По шагам, с командами для каждой платформы: Плагины Claude Code: маркетплейсы, установка и как собрать свой

Вопросы об ux-ui-agent-skills

Как установить ux-ui-agent-skills в Claude Code?

Сначала добавьте источник командой /plugin marketplace add plugin87/ux-ui-agent-skills, затем выполните /plugin install ux-ui-agent-skills@ux-ui-agent-skills.

Что делает ux-ui-agent-skills?

Он задаёт строгие правила проектирования UI и автоматически тестирует результат 52 проверками в реальном браузере.

Требуются ли плагину отдельные библиотеки или рантайм?

Нет, пакет работает без сборщиков и внешних зависимостей, передавая агенту готовые токены и правила.

Исходный текст

SKILL.md · MIT
<div align="center">

# UX/UI Agent Skills

*aka **XIAS***

### Design-system doctrine for coding agents, and **52 gates** that check what they build instead of taking their word for it.

Instructions, design tokens, 26 runnable skills and 2,144 design systems (2,087 of them measured from the live site) that make a coding agent design to a senior standard, in **any framework** and against **any design system**. Then the gates measure what it built.

<br>

<picture>
  <source media="(prefers-color-scheme: dark)" srcset=".github/brand/xias-cover-dark.png">
  <source media="(prefers-color-scheme: light)" srcset=".github/brand/xias-cover-light.png">
  <img src=".github/images/hero.png" alt="XIAS: design-system doctrine for coding agents. 52 objective gates, 34 of which open a real browser: WCAG contrast in every state, axe roles and landmarks, keyboard, target size, reduced motion, no horizontal overflow at 280, token by intent, screen economy. 2,144 design systems, 52 component specs, 26 runnable skills, 14 DTCG token files, 16 framework adapters. Claude Code, Codex, Cursor, Copilot, Gemini CLI, Antigravity, Aider, MCP." width="900">
</picture>

<br>
<br>

<a href="https://trendshift.io/repositories/24406" target="_blank"><img src="https://trendshift.io/api/badge/repositories/24406" alt="plugin87/ux-ui-agent-skills | Trendshift" width="250" height="55"></a>

<br>
<br>

[![Version](https://img.shields.io/badge/version-3.0.0-6366f1?style=for-the-badge)](https://github.com/plugin87/ux-ui-agent-skills/releases)
[![License: MIT](https://img.shields.io/badge/license-MIT-22c55e?style=for-the-badge)](#license)
[![WCAG 2.2 AA→AAA](https://img.shields.io/badge/WCAG-2.2_AA→AAA-a855f7?style=for-the-badge)](#-accessibility-standards)

<br>

[![npm](https://img.shields.io/npm/v/ux-ui-agent-skills?style=flat-square&logo=npm&logoColor=white&color=cb3837)](https://www.npmjs.com/package/ux-ui-agent-skills)
[![npm downloads](https://img.shields.io/npm/dt/ux-ui-agent-skills?style=flat-square&logo=npm&logoColor=white&color=cb3837)](https://www.npmjs.com/package/ux-ui-agent-skills)
![Tokens](https://img.shields.io/badge/Design_Tokens-DTCG-fbbf24?style=flat-square)
![Skills](https://img.shields.io/badge/runnable_skills-26-14b8a6?style=flat-square)
![Gates](https://img.shields.io/badge/objective_gates-52-16a34a?style=flat-square)
[![Live demo](https://img.shields.io/badge/live_demo-open-0ea5e9?style=flat-square)](https://plugin87.github.io/ux-ui-agent-skills/)
![Design Systems](https://img.shields.io/badge/design_systems-2144-f97316?style=flat-square)
![Frameworks](https://img.shields.io/badge/frameworks-any-8b5cf6?style=flat-square)
![Adapters](https://img.shields.io/badge/framework_adapters-16-22d3ee?style=flat-square)
![React 19](https://img.shields.io/badge/React-19-60a5fa?style=flat-square&logo=react)
![Next.js 15](https://img.shields.io/badge/Next.js-15-000000?style=flat-square&logo=nextdotjs)
![SwiftUI 6](https://img.shields.io/badge/SwiftUI-6-f472b6?style=flat-square&logo=swift)
![Tailwind v4](https://img.shields.io/badge/Tailwind-v4-38bdf8?style=flat-square&logo=tailwindcss)

</div>

**Current release [`v3.0.0`](https://github.com/plugin87/ux-ui-agent-skills/releases)** · [Changelog](CHANGELOG.md) · No build tools, dependencies, or runtime: a pure instruction and knowledge layer for AI agents.

---

## The output, rendered


Not a mockup. These are screenshots of the files in `examples/`, taken by
`node scripts/screenshot_docs.mjs` from the same HTML the 52 gates measure, so
what you see below is what the gate run passed, in both themes.

**Click through them yourself: [plugin87.github.io/ux-ui-agent-skills](https://plugin87.github.io/ux-ui-agent-skills/).**
There are 56 live pages (twenty whole-product screens, five websites from a
product launch to a Thai public service page, every component harness, both
reference screens and an iPhone Duo simulator), with a theme toggle. No install,
no clone.

<img src=".github/images/dashboard-light.png" alt="Atlas revenue console in light theme: sidebar navigation, a hero net-recurring-revenue figure at 2.48 million with a weekly bar chart, retention and churn cards, a donut of revenue by plan, an account ranking, regional sparklines, a sortable renewals table and an activity feed" />

<img src=".github/images/dashboard-dark.png" alt="The same Atlas console in dark theme, rendered from the same token theme with no per-page palette" />

<img src=".github/images/terminal-dark.png" alt="Meridian Terminal: a trading desk screen with a market ticker, candlestick chart with volume and moving average, order book with depth bars, cumulative depth curve, time and sales tape, watchlist sparklines, stacked area allocation, waterfall attribution, bubble chart and a correlation matrix" />

**Meridian Terminal** is the density test: **ten chart types on one screen**:
candlestick, volume, moving average, depth, order book, tape, sparklines,
stacked area, waterfall, bubble and a correlation matrix. Every one of them is
inline SVG drawn from the same tokens, with no chart library.
[Open it live.](https://plugin87.github.io/ux-ui-agent-skills/terminal/)
Build one yourself with **`/data-dashboard`**.

**Atlas** is one HTML file in `examples/showcase/`, built only from the kit's
tokens and rules: the hero figure leads at 3x the body size, the three
breakdowns are deliberately three different shapes rather than three identical
cards, the range tabs move the numbers, and the table headers really sort.
[Open it live.](https://plugin87.github.io/ux-ui-agent-skills/showcase/)

<table>
<tr>
<td width="50%"><img src=".github/images/reference-app-light.png" alt="Reference app in light theme: an Analytics screen led by one hero revenue metric, three smaller stats beneath it, a settings form, and a red Delete account action" /></td>
<td width="50%"><img src=".github/images/reference-app-dark.png" alt="The same Analytics screen in dark theme, rendered from the same token theme with no per-page palette" /></td>
</tr>
<tr>
<td><img src=".github/images/button-states-light.png" alt="Button harness in light theme showing primary, disabled, loading, secondary, danger and toggle states side by side" /></td>
<td><img src=".github/images/button-states-dark.png" alt="The same button states in dark theme, with the destructive action still wearing the danger variant" /></td>
</tr>
</table>

One theme, two modes, no per-page palette. The destructive action wears the
danger variant in both. Loading keeps full strength and swaps in a spinner
instead of borrowing the disabled dimming. Every one of those is a rule in
`CLAUDE.md` that a gate or a critic enforces.

---

## What the gates actually catch


Left: the same dashboard written the way a model writes it when nothing stops it:
the indigo-to-purple gradient, four equal cards with no focal point, emoji as
icons, one radius and one shadow everywhere, grey-on-white body text, and a blue
Delete Account. Right: the reference app in this repo.

<table>
<tr>
<td width="50%"><img src=".github/images/before-slop.png" alt="A generated-looking analytics dashboard: purple gradient header with emoji, four identical stat cards, grey low-contrast labels, tiny icon buttons, and a blue Delete Account button" /><br><b>Statistical defaults</b></td>
<td width="50%"><img src=".github/images/reference-app-light.png" alt="The same dashboard built to the kit's rules: one hero revenue metric leading, three smaller stats, a settings panel, and a red Delete account action" /><br><b>Built to the rules</b></td>
</tr>
</table>

The difference is not a matter of opinion, and that is the point. The page on the
left is `tests/fixtures/bad/slop-screen.html`; here is what the gates say about it:

| Gate | Verdict on the left-hand page |
|---|---|
| REAL-render WCAG | `x <h1> "Analytics Dashboard" 1.00:1 (need 3)`: white text on a gradient has no measurable background |
| State-aware WCAG | `x default "Save Changes" 1.00:1 (need 4.5) [rgb(255,255,255) on rgb(255,255,255)]`: 6 states below AA |
| Target size (2.5.8) | `x button.icon-btn is 15.3x16 (min 24x24)` |
| Responsive | `x @280px overflow +820px (widest: div.card)` |
| axe-core | `SERIOUS target-size` |
| Slop tells | HIGH: hardcoded indigo-purple gradient, single radius, one flat shadow, `#000` on `#fff` |
| Taste audit | HIGH: biggest heading 24px vs 14px body = 1.7x, not a display scale |
| Token by intent | `x "Delete Account" is destructive but filled with rgb(99, 102, 241) (hue 239deg, not a danger colour)` |
| No emoji | `x slop-screen.html:55: emoji/pictograph`: a chart glyph in the `<h1>` |
| No hardcoded values | `FAIL: 63 hardcoded value(s)` |

Ten gates reject it, none of them on a matter of taste. The right-hand page
passes all 52.

The last row is there because building this comparison broke a gate open.
`lint_intent` originally read that blue Delete Account as fine: it resolved
"primary" and "danger" from the page's own CSS variables, and a page with no
tokens resolved neither, so it skipped the page and reported zero intent-bearing
controls. An untokenised page is precisely where intent gets picked by
convenience, so the gate no longer looks away: a destructive label filled with a
saturated colour outside the danger hue range is wrong-intent with or without a
theme. Re-checked against every example in both themes afterwards: no false
positives.

`tests/meta/browser-gates.test.mjs` pins every claim in this table, so it cannot
rot.

---

## It was run blind, and it came back as rework

Two subagents were handed a brief and a project scaffolded by `ux-ui-skills new`,
which installs the kit and ships **no example screens**, and nothing else. No
hints, no warning that anything would be scored, no access to the conversation
that built the kit. They met `CLAUDE.md` and `.claude/rules/` the way a new
user's agent does.

**Both scored 14/14** on an independent run of `evals/run.mjs`, scored here
rather than self-reported. The rules transfer across a cold start.

**And the two runs cost the kit seven defects that four in-session runs never
hit**, because a familiar run keeps reaching for a finished demo theme instead of
the template a real user gets: a secondary button at **1.13:1** dark-on-dark
because the component tier never followed the dark map; a reduced-motion policy
in an external stylesheet read as "no policy" (Chromium treats a `file://` linked
sheet as cross-origin, so the gate was blind exactly where every real project
lives); a missing scrim token; a theme that emitted colours and no spacing; an
intent gate that did not know "cancel subscription" is destructive.

Then the outputs went to `/critique`, which renders the work and argues for
rejection. Verdict on two pages that pass all fourteen gates: **rework**, eight
findings, five Major, and not one of them measurable: an empty state that reads
as a page that failed to load, a theme toggle that communicates nothing about its
own state, a toast claiming "Draft project created" over a list that never
changed, and a loading state wearing the disabled dimming so it reads as "you
cannot do this".

**The blind outputs were never edited.** They are in `evals/out/` as the record
of what a cold-start agent produced; patching them would be editing the
experiment. The findings were spent on the kit instead. One of them became a
gate, `verify_interactive.mjs`, which fails any control that declares a state
contract and changes nothing when clicked.

The whole log, including what is still unproven, is in
[evals/RESULTS.md](evals/RESULTS.md).

---

## What It Does


| Capability | Description |
|-----------|-------------|
| **Design Token Generation** | Produces DTCG-format JSON tokens (colors, typography, spacing, shadows, borders, breakpoints, motion) with a 3-tier architecture: Primitive → Semantic → Component |
| **Component Design** | Designs components from Atoms to Templates following Atomic Design, with anatomy, variants, states, token mapping, and accessibility specs |
| **Code Generation (any framework)** | Adapter Protocol targets **any** stack (React+Tailwind, Next.js, SwiftUI, Vue, Svelte, Angular, Solid, Web Components/Lit, React Native, Flutter, Jetpack Compose, vanilla CSS, CSS-in-JS) or generates a new adapter on demand |
| **Design-System Interop** | Maps to/from **any** design system (Material 3, Apple HIG, Fluent, Carbon, shadcn/ui, Radix…) via a role-based crosswalk |
| **Runnable Skills** | 19 invocable `/skills` (each declaring `invocation: user|model`) + 5 slash commands + real scripts: token and contrast validators, real-render and state-aware WCAG gates, axe-core a11y, focus-trap, RTL, target size, keyboard, reduced motion, overflow, token-by-intent, taste and slop audits, token build |
| **Accessibility Auditing** | Evaluates against WCAG 2.2 AA/AAA with prioritized findings (P0/P1/P2) |
| **Design Review** | Scores designs across 6 dimensions with Nielsen's 10 Heuristics and a structured findings table |
| **Prototyping & Research** | Guides through a 5-level fidelity ladder, user journey mapping, and usability testing scripts |
| **Motion Design** | Tokenized durations, easing curves, transition presets, and reduced-motion strategy for accessible animation |
| **UX Writing** | Voice & tone system with error/empty-state formulas, microcopy patterns, and inclusive language guidelines |
| **Design Taste** | Native anti-slop doctrine, aesthetic archetypes, and a library of **2,144 design systems** for layout variance, editorial typography, and premium visual direction |

---

## Quick Start


> **Five ways in, and which one is yours:** [docs/INSTALL.md](docs/INSTALL.md)
> walks through the Claude Code plugin, the `AGENTS.md` surface for Codex and
> Cursor, the MCP server, Homebrew and `npx`: what each one installs, and what
> you give up by choosing it.

### Option A: Install as a Claude Code plugin (recommended)

Two lines in Claude Code, and every skill, command, and agent is available in any
project you open, with no files copied into your repo:

```
/plugin marketplace add plugin87/ux-ui-agent-skills
/plugin install ux-ui-agent-skills@ux-ui-agent-skills
```

You get 26 skills and the `design-critic` agent.

**Fourteen the model reaches for on its own** when the work calls for them:
`design-tokens`, `design-component`, `design-code`, `design-review`,
`a11y-audit`, `apply-aesthetic`, `data-dashboard`, `design-qa`,
`figma-integration`, `performance`, `token-build`, `ux-writing`,
`design-doctrine`, and `design-screen`, the one that claims "design a page /
screen / app", which until 2026-10-05 no skill did, so the most common request
there is matched nothing strongly and the kit got used at a fraction of its
depth.

**Twelve you start yourself**, because each one takes an action or sets a
direction that should be your call: `/brandkit`, `/governance`, `/image-to-code`,
`/migrate-design-system`, `/prototype`, `/redesign`, `/gate`, `/critique`,
`/grill-me`, `/ship`, `/scaffold-project`, `/observe`. They carry
`disable-model-invocation: true`, which is the field that actually stops the
model invoking them. Until 2026-10-05 six of them said `invocation: user`, a
key Claude Code does not read, so they were auto-invocable the whole time. The
last five were commands and are skills now, so they have a `CLAUDE_SKILL_DIR` of
their own and their paths resolve under a plugin install.

`design-doctrine` carries the house rules that `CLAUDE.md` carries in the repo,
because a plugin root `CLAUDE.md` is not loaded as project context, as Claude
Code's own `plugin validate` says so.

**Then just work.** Ask for the thing you want and the right skill loads itself:

```text
"Design a notification component with all states and accessibility"
"Build the billing settings screen, one shared theme, light and dark"
"/grill-me"      interrogate the brief before anything is built
"/gate"          run all 52 checks and report the real N/N
"/critique"      hand the result to a critic that argues for rejection
"/observe"       did the agent follow the rules? sessions, blocks, findings left open
```

If the skills do not show up straight away, start a new session. To check what
is loaded, update, or remove it:

```bash
claude plugin details ux-ui-agent-skills    # inventory + token cost per skill
claude plugin update  ux-ui-agent-skills
claude plugin uninstall ux-ui-agent-skills
claude plugin marketplace remove ux-ui-agent-skills
```

`uninstall` also deletes the plugin's data folder, which holds the gate
receipts and the `/observe` history.

**34 of the gates open a real browser.** They use the Chrome installed on the
machine, or the one `npx playwright install chrome` downloads. Without either
they cannot measure: run through `scripts/gate.mjs`, as the skills do, they fail
with `REQUIRED, FAILING` instead of passing on nothing.

### See it first, install nothing

```bash
npx ux-ui-agent-skills demo        # copies the rendered examples and opens them
```

Every page it opens is a page the gates measure. Delete the folder afterwards;
nothing was installed. Or skip the copy entirely and use the
[live demo](https://plugin87.github.io/ux-ui-agent-skills/).

### Option B: Use it from any MCP client

```bash
claude mcp add ux-ui-gates -- npx -y --package=ux-ui-agent-skills ux-ui-mcp
```

Six tools: `list_gates`, `run_gate`, `review_ui`, `get_doctrine`,
`get_design_system`, `get_tokens`. Read the doctrine and the tokens before
generating, run the gates on what you produced after.

`AGENTS.md` already lets any agent **read** the doctrine. This lets any MCP
client **run the 52 gates**, which reading cannot do. Plenty of things can
describe good UI, very little can tell you afterwards that the contrast you
shipped is 3.9:1 on hover.

It returns what the gates printed, unedited, with their exit codes: 0 looked and
found nothing, 1 found something, **2 could not look**, never a pass. No SDK
dependency; the protocol is implemented directly so `"dependencies": {}` stays
true. Details in [docs/MCP.md](docs/MCP.md).

### Option C: Install with Homebrew

```bash
brew install plugin87/tap/ux-ui-agent-skills
ux-ui-skills init                   # CLAUDE.md + .claude/ skills, rules, hooks
ux-ui-skills init --agent codex     # AGENTS.md, for Codex, Cursor, Copilot, Aider
```

Same npm tarball the `npx` route runs, so this is a convenience rather than a
second product that can drift. The formula is generated from the registry and
its `test do` block asserts the installed CLI reports the version the formula
names, and that `--agent codex` writes `AGENTS.md` without `CLAUDE.md` or
`.claude/`. Details in [docs/HOMEBREW.md](docs/HOMEBREW.md).

### Option D: Install with `npx`

Drop the kit into any project, no clone needed:

```bash
npx ux-ui-agent-skills init          # full kit into the current folder
npx ux-ui-agent-skills add tokens taste design-systems   # just some areas
npx ux-ui-agent-skills list          # see all areas
npx ux-ui-agent-skills library get stripe   # fetch one design-system spec
```

Flags: `--force` (overwrite existing files) · `--dry` (preview, change nothing).

The 2,144 design-system specs are 27.6MB, and a project uses one or two, so
`init` does not copy them. It installs an index of every spec with its sha256,
and the agent fetches the one it picks with `node scripts/library.mjs get
<name>`. A fetched file whose sha256 does not match the index is refused and
nothing is written. `library pull` fetches all of them for offline work. The
plugin (Option A) and a clone carry every spec already. To browse all 2,144, open
[design-systems/catalog.md](design-systems/catalog.md): GitHub's folder view lists
only the first 1,000.

Working on the kit itself, or want it vendored? [Clone and copy](docs/GUIDE.md#install-from-a-clone) instead.

**Then start using:** open the project in **Claude Code** or any Claude-powered IDE. `CLAUDE.md` loads automatically, activating the agent persona with full access to every tokens / components / taste / design-system file and the runnable `/skills`.

<details>
<summary><b>Example prompts</b></summary>

<br>

```text
"Design a notification component with all states and accessibility"
"Review this login page against WCAG 2.2 and Nielsen's heuristics"
"Generate React + Tailwind code for a data table with sorting and pagination"
"Create a color token palette for a fintech brand using blue as the primary"
"Audit this form for accessibility issues, and give me a prioritized findings table"
"Write the empty state and error copy for the onboarding flow"
"Spec the motion for the modal open/close with reduced-motion fallback"
```

</details>

---

## Proving It, and Admitting What Cannot Be Proven


The kit ships **52 objective gates** behind one command:

```bash
node scripts/accuracy_report.mjs     # 52/52 or it fails, no partial credit
```

**34 of them open a real browser, so they need one installed.** `npx
ux-ui-agent-skills init` does not install Playwright. Whether `/plugin install`
does depends on the Claude Code version: 2.1.301 installed the kit's npm
dependencies, Playwright included, into the plugin folder. Neither installs the
browser itself. In a clone, run this once before expecting a full score:

```bash
npm install                          # playwright
npx playwright install chrome        # real Chrome: six gates require the channel
```

Without it those 34 report `REQUIRED, FAILING` under `accuracy_report.mjs`, which
is the honest answer. Run individually they print `SKIPPED` and **exit 0**, so
run a single render gate through `scripts/gate.mjs` (or prefix it with
`DS_REQUIRE_BROWSER=1`) if you are reading its exit code, rather than reading
silence as green. Every skill already does.

Token validity, WCAG contrast on a real headless render in light *and* dark, every
element in default/hover/focus, axe roles and names, focus traps, RTL, responsive
from 280 to 951px and folded on iPhone Duo, target size, keyboard operability, reduced motion (including
content that only an animation reveals), silent text clipping, token-by-intent,
and zero emoji anywhere in the output or the instruction surface.

**What that number covers, stated exactly.** 34 of the 52 checks open a real
browser, so what they measure is **rendered HTML**: the 23 component harnesses,
the twenty industry screens, the reference app, the live demo, the starter
template, and the React source compiled and server-rendered. The other 18 read
files: token JSON and alias resolution, contrast math on the token source,
component specs, hardcoded values, theme references, emoji, hooks, the
instruction surface held in step across `CLAUDE.md` and `AGENTS.md`, and
destructive-intent declarations in framework source.

The split is not typed here: `tests/meta/counts.test.mjs` derives it from which
gates import Playwright and fails if this paragraph drifts from the report. It
read 35/16 until 2026-10-07, when the derivation was first run. Those were two numbers
nobody had measured, in the section about not stating numbers you did not
measure.

Framework source (`.tsx`, `.vue`, `.swift`) gets the file-reading checks: no
emoji, no hardcoded values, every `var(--…)` resolving to the theme, and, since a
blue Delete shipped in this repo's own `Settings.tsx` while the HTML twin of that
screen was correct, every destructive control declaring its intent
(`lint_intent_source.mjs`). That last one proves a **declaration**, never a
colour: `variant="destructive"` wired to a blue token passes it.

Only a render catches that, so the React components are now compiled with esbuild,
server-rendered, and put through the same browser gates as the HTML
(`render_framework_source.mjs`, which also fails if any class the components emit
resolves in no linked stylesheet). `.vue` and `.swift` are still file-checks only,
and are not proven by this number until they are rendered and measured.

That is correctness. It is not quality, and the kit says so out loud:

| Question | Answer | How |
|---|---|---|
| Is it correct? | Measured, all or nothing | `node scripts/accuracy_report.mjs` -> a real `N/N` |
| Is it any good? | Judged, never scored | `/critique`: an adversarial `design-critic` that renders the work, argues for rejection, and cites evidence per finding |
| Did the agent follow the rules? | Counted per session, from the hooks | `/observe` or `npx ux-ui-skills observe` -> pages edited and measured, Stop hook blocks, findings left open, sessions that quit while blocked |
| Does the kit transfer to a cold start? | Measured, one brief at a time | `evals/`: cold-start briefs, then `node evals/run.mjs <brief-id>` points 15 objective gates at what the agent produced |

`/critique` exists because a passing gate is never evidence of taste. It refuses to
review from source alone, screenshots at 1280 and 390 in both themes, clicks every
control, and returns a verdict with the three reasons a senior designer would send
the work back.

It runs as a **forked subagent**, so the critic never sees the conversation that
produced the work. That is not a performance decision. A critic holding the
maker's reasoning is reviewing the argument instead of the artifact, and it
agrees far too easily.

The critic itself is scored. `tests/fixtures/critic/` holds four pages carrying
fourteen seeded design defects (no focal point, an empty state with no way
forward, a toast that claims a project was created while the list is unchanged),
and **every one of those pages passes all 52 gates**. That is the property that
makes the number mean something: nothing in the harness can reach these, so
catching them takes judgement. `node evals/score_critic.mjs` scores a critique
against the key, prints the matching excerpt for every hit and the rule for every
miss, and says in its own output that the number is a floor until you have read
them.

`/observe` answers a different question: not whether the work is right, but
whether the agent did what the rules say while making it. The hooks record one
event per session start, per UI edit (with what each fast gate caught) and per
end of turn (which pages had a passing gate, and whether the Stop hook passed,
blocked, or let the agent go with pages still open). The report joins those to
the gate receipts and gives each session a verdict from three counted rules:
every edited page measured, no finding left on a file's latest version, never
finished while blocked. `--last` and `--session <id>` print a timeline;
`--html` writes the same thing as a page. Everything stays on the machine,
files older than 30 days are deleted, and `DS_OBSERVE=0` turns it off.

Runs are recorded in `evals/RESULTS.md` with their provenance attached: who built the output and whether they could see the kit while doing it, because a run without that context is not evidence of anything.

The eval suite exists because "the kit's own examples pass" is a weaker claim than
"an agent given only this kit and a brief produces work that passes". Building it
caught two real defects the 34-check gate had missed. See `evals/README.md`.

---

## Documentation

| Where | What is in it |
|---|---|
| **[Live demo](https://plugin87.github.io/ux-ui-agent-skills/)** | 49 rendered pages: twenty industry screens, every component harness, both reference screens, a theme toggle |
| [docs/GUIDE.md](docs/GUIDE.md) | Using it as a plugin (inventory, management, token cost), how the skills compose, the repo map, token architecture, frameworks, interop, a11y standards, starting a new product project |
| [CHANGELOG.md](CHANGELOG.md) | Every release, newest first |
| [docs/BRANCHING.md](docs/BRANCHING.md) | One concern per branch, one commit per merge, and the settings that enforce it |
| [CONTRIBUTING.md](CONTRIBUTING.md) | Why PRs are not merged, what to report instead, and how a gate that can still say no is built |
| [CLAUDE.md](CLAUDE.md) | The always-on brief the agent actually reads |
| [.claude/rules/](.claude/rules/) | The depth behind it: tokens and colour, type and spacing, components, accessibility, frameworks, review, brand and operations |
| [taste/](taste/) | The anti-slop doctrine and motion choreography. The 2,144 design systems live in [design-systems/](design-systems/) |

---

## Not using Claude Code?

The kit works with **Codex, Cursor, Copilot, Jules, Aider, VS Code and about
twenty more**, through the open `AGENTS.md` convention, and with **Gemini CLI
and Antigravity**, which read `GEMINI.md` and `.agents/skills/`.

```bash
npx ux-ui-agent-skills init --agent codex    # AGENTS.md + the whole kit
npx ux-ui-agent-skills init --agent gemini   # GEMINI.md + 23 ported skills: Gemini CLI, Antigravity
npx ux-ui-agent-skills init --agent both     # CLAUDE.md and AGENTS.md
npx ux-ui-agent-skills init                  # Claude Code (default)
gemini extensions install https://github.com/plugin87/ux-ui-agent-skills   # Gemini CLI, every project
```

What transfers, and what does not, stated exactly:

| | Claude Code | AGENTS.md surface | Gemini CLI, Antigravity |
|---|---|---|---|
| Tokens, components, taste, accessibility, 2,144 design systems | yes | **yes** | **yes** |
| All 52 gates | yes | **yes**, they are plain Node and Python | **yes** |
| Doctrine: the eight states, token by intent, mobile-first, the house rules | yes | **yes** | **yes**, in `GEMINI.md` |
| 26 skills that load on demand | yes | no, the doctrine is inlined instead | 23 ported skills |
| Hooks that fire whether or not the model remembers | yes | **no** | **no** |
| `/critique` in a separate context | yes | no, it is a step in the instructions | no |

A skill is ported, not copied: Gemini CLI and Antigravity do not expand
`${CLAUDE_SKILL_DIR}`, so each path is rewritten for where the skill lands, and
the three that need Claude Code (`critique`, `observe`, `scaffold-project`) stay
there. Setup and what was measured: [docs/INSTALL.md](docs/INSTALL.md#2b-gemini-cli-and-antigravity).

**The hooks are the real difference.** On Claude Code...

© Thientan Soparat · MIT · plugin87/ux-ui-agent-skills

Похожие плагины

Все плагины по теме «Фронтенд и UI» →
design-visual Rudolf S. Субагент design-visual для выстраивания дизайн-систем, дизайн-токенов и библиотек компонентов в Claude Code с проверкой доступности по стандарту WCAG 2.1 AA. CX Codex ★ 587звёзд репо ux-designer szilu Скилл ux-designer для Claude Code помогает проектировать интерфейсы и проверять готовые экраны или frontend-код. Включается при работе с компонентами, доступностью, формами, навигацией и микрокопирайтом. CX Codex +1 1 900установок glassmorphism typeui.sh Скилл glassmorphism для Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI и OpenCode: задаёт правила стеклянных интерфейсов с полупрозрачными слоями и светящимися границами. Генерирует готовые к реализации гайдлайны, а не общие советы. CX Codex +1 3 500установок pulse typeui.sh Задаёт правила дерзкой дизайн-системы Pulse с рамками 4px, яркой оранжевой палитрой и шрифтом Limelight. Помогает писать дизайн-гайды и проверять компоненты на WCAG 2.2 AA. CX Codex +1 2 400установок generate-ui-from-brand dembrandt Скилл generate-ui-from-brand превращает URL сайта или файл DESIGN.md в дизайн-токены, решения по UX и готовую спецификацию интерфейса. CX Codex +1 1 000установок color-system Owl-Listener Скилл color-system для Claude Code собирает палитру продукта: тональные шкалы, семантические роли и проверку контраста по WCAG. Помогает при создании или переделке цветовой системы с нуля. CX Codex +1 3 000установок

Из того же набора35

Весь набор →
critique Скилл critique для Claude Code разбирает готовый интерфейс и ищет причины отправить его на доработку. Запускается после прохождения проверок проекта. CX Codex +1 ★ 3 216звёзд репо design-critic Субагент design-critic для Claude Code оценивает интерфейсы с позиции дизайн-директора и ищет поводы отклонить макет. Запускается после прохождения тестов на готовых экранах. CX Codex ★ 3 216звёзд репо grill-me Скилл grill-me для Claude Code задаёт вопросы о проекте до начала разработки и записывает оставшиеся решения как допущения. Его запускают перед созданием интерфейса. CX Codex +1 ★ 3 216звёзд репо tokens-and-color Правило tokens-and-color для Claude Code подключается при создании, расширении или проверке токенов, настройке темы и выборе цвета. Оно задаёт уровни токенов и требования к контрасту. CX Codex +1 ★ 3 216звёзд репо design-screen Скилл design-screen для Claude Code проектирует и собирает экраны, сайты и интерфейсы по запросу. Для дашбордов, редизайна и вёрстки по скриншоту он передаёт задачу профильным скиллам. CX Codex +1 ★ 3 216звёзд репо design-doctrine Скилл design-doctrine для Claude Code задаёт правила вёрстки и запускает скрипты проверки интерфейса. Включается перед сборкой, ревью компонентов или настройкой дизайн-токенов. CX Codex +1 ★ 3 216звёзд репо data-dashboard Скилл data-dashboard для Claude Code проектирует плотные аналитические экраны с графиками и таблицами. Включается для дашбордов, торговых терминалов и мониторинговых панелей. CX Codex +1 ★ 3 216звёзд репо components Правило components загружается при проектировании, уточнении или рецензии компонента. Содержит принципы атомарного дизайна, композиции и интервалов — то, что не видно ни в каких гейтах, но видно в коде сразу. CX Codex +1 ★ 3 216звёзд репо

Спонсоры проекта

Компании, которые помогают держать аналитику открытой