Eval Ui Ux logo

Eval Ui Ux

Community
Ibrahim-3d
eval-ui-ux

Specialized UI/UX evaluator for the Evaluate-Loop. Use this for evaluating UI shell tracks, design system tracks, screen implementation tracks, or any track where the primary deliverable is visual/interactive UI. Checks design system adherence, visual consistency, layout structure, responsive behavior, component states, animations, accessibility baseline, and usability check (copy quality). Dispatched by loop-execution-evaluator when track type is 'ui', 'design-system', or 'screens'. Triggered by: 'evaluate UI', 'UI review', 'design review', 'visual audit'.

Overview

PublisherIbrahim-3d
Repositoryorchestrator-supaconductor
Skill nameeval-ui-ux
Stars
378
Forks
38
Bundled files
Instructions only
LicenseAGPL-3.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by Ibrahim-3d on GitHub. Read the source before you install it.

Installation

Install the Eval Ui Ux AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Ibrahim-3d/orchestrator-supaconductor.git /tmp/orchestrator-supaconductor
mkdir -p .claude/skills
cp -r /tmp/orchestrator-supaconductor/skills/eval-ui-ux .claude/skills/eval-ui-ux
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Eval Ui Ux in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Eval Ui Ux on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Eval Ui Ux is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

UI/UX Evaluator Agent

Specialized evaluator for tracks whose deliverables are visual UI — screens, components, design systems, layouts.

When This Evaluator Is Used

Dispatched by loop-execution-evaluator when the track is one of:

  • Screen implementation
  • Design system work
  • Component library
  • UI polish or UX audit

Inputs Required

  1. Track's spec.md — what was supposed to be built
  2. Track's plan.md — tasks that should be complete
  3. Design system reference — your project's global CSS or token file (e.g., src/app/globals.css)
  4. Components to evaluate — all files in src/components/ and src/app/
  5. Data files — content JSON files used for copy (if applicable)

Evaluation Passes (8 checks)

Pass 1: Design System Adherence

read_file your project's CSS/token file to extract the token system, then check components:

CheckWhat to Look For
ColorsCSS custom properties (--color-*, --brand-*) used, no raw hex/rgb in components
SpacingTailwind spacing classes follow consistent grid, no arbitrary px values
TypographyFont families from your design system fonts, sizes from scale
RadiusUses token-defined radius values, no random rounded-* overrides
ShadowsShadow classes from token system, consistent elevation levels
Glass-morphismBackdrop-blur, bg-opacity patterns on cards/modals/overlays (if applicable)
markdown
### Design System Adherence: PASS / FAIL
- Hardcoded colors found: [count] — [list files:lines]
- Hardcoded spacing found: [count] — [list files:lines]
- Typography violations: [count] — [list]
- Token coverage: [X]% of visual properties use design tokens

Pass 2: Visual Consistency

Compare styling patterns across screens:

CheckWhat to Look For
Spacing rhythmSame gap/padding patterns across sections
Color usageBrand palette applied consistently (not random grays)
Component stylingSame component (Card, Button) looks identical on all pages
Icon sizingIcons use consistent size props
Page structureSimilar content types have similar visual treatment
markdown
### Visual Consistency: PASS / FAIL
- Inconsistencies found: [count]
- Affected screens: [list]
- Specific issues: [describe each]

Pass 3: Layout & Structure

CheckWhat to Look For
Header presenceHeader component rendered on every page (or layout group)
Footer presenceFooter component rendered on every page
Container usageMax-width Container wraps content on all pages
Section usageVertical spacing via Section component
Visual hierarchyh1 → h2 → body → actions ordering clear
Content widthNo full-bleed text blocks (constrained width)
markdown
### Layout & Structure: PASS / FAIL
- Pages missing Header: [list]
- Pages missing Footer: [list]
- Pages missing Container: [list]
- Hierarchy issues: [describe]

Pass 4: Responsive Behavior

Check component classes and layout patterns:

BreakpointWhat to Check
375px (mobile)Single column, stacked layout, touch-friendly
768px (tablet)2-column grids, adjusted spacing
1024px+ (desktop)Full layout, 3-4 column grids
CheckWhat to Look For
Grid collapsegrid-cols-1 md:grid-cols-2 lg:grid-cols-3 patterns
Horizontal scrollNo overflow-x issues, no fixed-width elements
Touch targetsButtons/links >= 44px on mobile (min-h-11, p-3, etc.)
Mobile menuHeader collapses to hamburger/sheet on mobile
Text truncationLong text doesn't break layout
markdown
### Responsive: PASS / FAIL
- Breakpoints covered: [375/768/1024]
- Pages with issues: [list]
- Touch target violations: [list components]

Pass 5: Component States & Conditional Rendering

Check interactive components for complete state coverage and safe rendering logic:

Component TypeRequired States
Buttondefault, hover, active, focus, disabled, loading
Inputdefault, focus, error, disabled, placeholder
Carddefault, hover (if interactive)
Modalopen/close animation, backdrop, focus trap, escape-to-close
Toastsuccess, error, info variants
Loadingspinner or skeleton for every async operation

Conditional Rendering Patterns to Check:

Anti-PatternWhat to FlagFix
Magic string comparisonstatus === 'ready' && <Image />Use explicit arrays: ['ready', 'locked'].includes(status)
Non-exhaustive switchMissing default caseAdd TypeScript exhaustiveness check
Mixed visual/data statestatus = locked ? 'locked' : 'ready'Separate: const isLocked = item.locked; const hasContent = !!imageUrl;
Status explosion7+ status codesSplit into orthogonal states

Example Brittle Pattern:

typescript
// BAD: Image disappears when status changes
{imageUrl && status === 'ready' && <Image />}
// When item is locked, status becomes 'locked' → image hidden!

// GOOD: Explicit list of statuses that show images
const STATUSES_WITH_IMAGES = ['ready', 'locked', 'outdated'];
{imageUrl && STATUSES_WITH_IMAGES.includes(status) && <Image />}

// BETTER: Separate concerns
const shouldShowImage = imageUrl && !['generating', 'error'].includes(status);
{shouldShowImage && <Image />}
markdown
### Component States & Rendering: PASS / FAIL
- Missing states: [component: missing state]
- Components audited: [count]
- Coverage: [X]% have all required states
- **Brittle conditionals found: [count] — [list files:lines]**
- **Non-exhaustive switches: [count] — [list]**
- **Status explosion (>7 codes): [list components]**

Pass 6: Animation & Transitions

CheckWhat to Look For
Page transitionsPage transition wrapper or framer-motion layout
Hover effectsSubtle scale/shadow/opacity changes on interactive elements
Loading animationsSpinner/skeleton with smooth animation
Modal transitionsFade/scale on open/close
State transitionsNo sudden jumps between states
markdown
### Animations: PASS / FAIL
- Pages missing transitions: [list]
- Components missing hover effects: [list]
- Jarring state changes: [describe]

Pass 7: Accessibility Baseline

CheckWCAG LevelWhat to Look For
LabelsAAll <input> elements have associated <label>
Button textAAll buttons have visible text or aria-label
Alt textAAll <img> elements have meaningful alt
Color contrastAAText/background contrast >= 4.5:1 (body), >= 3:1 (large)
Focus visibleAAFocus ring visible on all interactive elements
Focus orderATab order follows visual reading order
Modal focusAModal traps focus when open
markdown
### Accessibility: PASS / FAIL
- Missing labels: [list inputs]
- Missing alt text: [list images]
- Contrast issues: [list]
- Focus order issues: [describe]

Pass 8: Usability Check (Copy Quality)

read_file all user-facing text in components and data files:

CheckWhat to Look For
HeadingsSimple, friendly, no jargon
CTAsAction verbs that clearly describe the outcome
ErrorsTell user what to do, not technical details
LabelsEveryday language (e.g., "Your name", not "Name identifier")
JargonNo: "auth", "render", "deps", "schema", "API", "submit"
ToneReassuring, not pushy (especially paywall, error states)

Verify against your target personas — would they understand every piece of text without explanation?

markdown
### Usability Check: PASS / FAIL
- Jargon found: [word: file:line]
- Unfriendly copy: [text: file:line]
- Suggested rewrites: [original → suggested]

Verdict Template

markdown
## UI/UX Evaluation Report

**Track**: [track-id]
**Evaluator**: eval-ui-ux
**Date**: [YYYY-MM-DD]
**Screens Evaluated**: [count]

### Results
| Pass | Status | Issues |
|------|--------|--------|
| 1. Design System | PASS/FAIL | [count] issues |
| 2. Visual Consistency | PASS/FAIL | [count] issues |
| 3. Layout & Structure | PASS/FAIL | [count] issues |
| 4. Responsive | PASS/FAIL | [count] issues |
| 5. Component States | PASS/FAIL | [count] issues |
| 6. Animations | PASS/FAIL | [count] issues |
| 7. Accessibility | PASS/FAIL | [count] issues |
| 8. Usability Check | PASS/FAIL | [count] issues |

### Verdict: PASS / FAIL
[If FAIL, list specific fix actions for loop-fixer]

Handoff

  • PASS → Return to loop-execution-evaluator → Conductor marks complete
  • FAIL → Return to loop-execution-evaluator → Conductor dispatches loop-fixer

Frequently asked questions

What does the Eval Ui Ux AI skill do?

Specialized UI/UX evaluator for the Evaluate-Loop. Use this for evaluating UI shell tracks, design system tracks, screen implementation tracks, or any track where the primary deliverable is visual/interactive UI. Checks design system adherence, visual consistency, layout structure, responsive behavior, component states, animations, accessibility baseline, and usability check (copy quality). Dispatched by loop-execution-evaluator when track type is 'ui', 'design-system', or 'screens'. Triggered by: 'evaluate UI', 'UI review', 'design review', 'visual audit'.

Why use Eval Ui Ux on TypingMind?

Because you install it once and use it with any model. Eval Ui Ux is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Eval Ui Ux in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Ibrahim-3d/orchestrator-supaconductor/tree/master/skills/eval-ui-ux. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Eval Ui Ux?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Eval Ui Ux?

As many as you like. As long as a model supports skills, you can use Eval Ui Ux with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Eval Ui Ux AI skill free?

Yes. It is published on GitHub by Ibrahim-3d under the AGPL-3.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇