Synthèse automatique
Cette skill effectue une revue UX orientée efficacité pour des outils de productivité et des interfaces d'administration, en évaluant la navigation clavier, la palette de commandes, l'édition inline et l'UI optimiste.
Points forts
- Grille d'évaluation structurée par niveaux (manquant, basique, bon, excellent) pour chaque motif UX
- Analyse du code via quatre sous-agents Explore lancés en parallèle (palette, motifs UI, navigation, feedback)
- Exploite les documents de spécification (.specs/) en plus du code pour recouper les intentions et l'implémentation
- Principes explicites et mémorables, comme « les mains restent sur le clavier »
Limites
- Orientée exclusivement vers les outils pro et back-offices : peu pertinente pour des sites vitrines ou des parcours grand public
- Suppose que la cible soit définie (spec, chemins ou diff de branche) et interroge l'utilisateur sinon
- Fournit une grille de notation mais pas de correctifs concrets ni de code de remédiation
Utilisez-la pour auditer une fonctionnalité existante d'un outil de productivité et vérifier qu'elle procure une expérience rapide et professionnelle.
Ne l'utilisez pas pour évaluer l'accessibilité générale, le design visuel de marque ou des interfaces grand public où le clavier n'est pas central.
Analyse de sécurité
SûrThe skill only instructs reading repository documents, analyzing code via read-only subagents, and evaluating UX patterns. It does not declare or use any tools that could execute commands, modify files, or access network resources. No destructive, exfiltrating, or obfuscated actions are present.
Aucun point d'attention détecté
Comment cette analyse est faiteExemples
Review the UX of the .specs/bulk-edit/ spec — check whether the bulk editing flow is keyboard-first and uses inline editing instead of modals.Do a UX review of all changes on this branch vs main, focusing on command palette coverage and optimistic UI for the new task list.review:ux src/components/TaskList.tsx src/components/TaskDrawer.tsx — does this feel like a pro tool with proper keyboard navigation?name: review:ux description: UX Review - analyzes feature for efficiency-first UX patterns, keyboard navigation, and pro-tool experience argument-hint: <spec-name|path|branch>
UX Review
Performs a UX review optimized for productivity tools and admin interfaces. Based on the principle: "Hands stay on keyboard, mouse is optional."
This review evaluates features against efficiency-first UX patterns used in best-in-class productivity tools: command palettes, keyboard-first navigation, inline editing, optimistic UI, and minimal context switching.
When to use
Use this skill when the user needs to:
- Validate that a feature feels fast and professional
- Check if keyboard shortcuts and command palette are properly integrated
- Verify inline editing vs unnecessary page transitions
- Ensure optimistic UI and instant feedback
- Review a feature for "pro tool" experience quality
Core UX Principles
The interface should think faster than the user
- Command Palette — single entry point for all actions (⌘K / Ctrl+K)
- Keyboard-First — frequent actions have shortcuts, arrow navigation works
- Inline Editing — no modals for simple edits, Enter saves, Esc cancels
- Minimal States — side panels over pages, overlays without context loss
- Search-First — search is the primary navigation method
- Strong Defaults — zero decisions needed for common actions
- Optimistic UI — instant response, no spinners when avoidable
- List-Centric — lists as main screens, saved views, fast filters
- Context > Settings — actions next to objects, not buried in settings
- Visual Clarity — no noise, color = meaning only
Instructions
Step 1: Determine Review Target
Parse $0 to determine what to review:
- Spec name — if provided, read documents from
.specs/<spec-name>/(requirements, design, tasks) - File paths — if provided, review specific files or directories
- Branch diff — if
branchis specified, review all changes on the current branch vs main
If no arguments provided:
- Check if there are specs in
.specs/and list them - Check if there are uncommitted changes or branch changes
- Ask the user what to review
Step 2: Gather Context
Read all available materials depending on the target:
From spec documents (if available):
requirements.md— extract user stories, acceptance criteria, and described user flowsdesign.md— extract UI components, data flow diagrams, and interaction patterns
From codebase:
Use parallel sub-agents (subagent_type: "Explore") to investigate:
- Command palette agent — find command palette implementation, registered commands, keyboard shortcut definitions
- UI patterns agent — find forms, dialogs, pages, lists; identify inline editing vs modal patterns
- Navigation agent — find route definitions, side panels, drawers, overlay patterns
- Feedback agent — find optimistic updates, loading states, toast notifications, undo mechanisms
Launch all four agents in a single message (parallel tool calls).
Step 3: Evaluate Efficiency-First Patterns
For each pattern, evaluate the current implementation level:
3a. Command & Control
| Pattern | ❌ Missing | ⚠️ Basic | ✅ Good | 🚀 Excellent | |---------|-----------|----------|---------|--------------| | Command Palette | No ⌘K | Opens but limited actions | Search + navigation + actions | Context-aware commands, recent items, fuzzy search | | Keyboard Shortcuts | None | Only global (save, close) | CRUD shortcuts (C/E/D) + navigation (↑↓) | Full coverage, discoverable hints, customizable | | Focus Management | Random focus | Focus on page load | Logical tab order | Keyboard trap-free, skip links, roving tabindex |
3b. Editing & Data Entry
| Pattern | ❌ Missing | ⚠️ Basic | ✅ Good | 🚀 Excellent | |---------|-----------|----------|---------|--------------| | Inline Editing | Always navigate away | Modal for every edit | Click-to-edit in lists | Enter=save, Esc=cancel, blur=save, multi-field inline | | Strong Defaults | All fields required | Some defaults | Smart defaults based on context | Auto-fill from context, minimal required fields | | Form Efficiency | One field per page | Standard form | Logical grouping, clear optionals | Auto-advance, paste-to-fill, batch entry |
3c. Navigation & Context
| Pattern | ❌ Missing | ⚠️ Basic | ✅ Good | 🚀 Excellent | |---------|-----------|----------|---------|--------------| | Minimal States | Full page transitions | Modals for everything | Side panels for details | Overlays preserve context, split views, peek preview | | Search-First | No search | Basic text match | Instant search, multiple entities | Filters, saved searches, search by ID/status/user | | List-Centric | Card grids only | Basic lists | Sortable, filterable lists | Saved views, column customization, bulk actions |
3d. Feedback & Performance
| Pattern | ❌ Missing | ⚠️ Basic | ✅ Good | 🚀 Excellent | |---------|-----------|----------|---------|--------------| | Optimistic UI | Wait for server | Spinner on every action | Optimistic for reads | Full optimistic with rollback, background sync | | Response Time | Visible delays | < 500ms responses | < 200ms, skeleton loading | < 100ms, no spinners, prefetching | | Undo/Redo | Confirm dialogs only | Toast with undo for delete | Undo for all destructive | Full undo stack, Ctrl+Z support |
3e. Visual Design
| Pattern | ❌ Missing | ⚠️ Basic | ✅ Good | 🚀 Excellent | |---------|-----------|----------|---------|--------------| | Visual Clarity | Cluttered, many icons | Some organization | Clean, purposeful layout | Typography-driven, color = status only | | Context > Settings | Actions in settings page | Actions in dropdown menu | Actions visible on hover | Inline actions, contextual toolbars | | Information Density | Too sparse or too dense | Acceptable density | Scannable, grouped info | Progressive disclosure, smart truncation |
Step 4: Cross-Reference with Spec
If spec documents are available, verify:
- Keyboard flows documented — spec mentions keyboard shortcuts and navigation
- Inline editing specified — design prefers inline over modal where appropriate
- Performance requirements — response time and optimistic UI mentioned
- Command palette integration — new actions registered in command palette
Step 5: Generate UX Review Report
Present a structured report:
# UX Review: [Feature Name]
**Scope:** [What was reviewed — spec docs, files, branch diff]
**UX Maturity:** [🚀 Pro-tool / ✅ Good / ⚠️ Basic / ❌ Needs Work]
---
## Efficiency-First Assessment
### Command & Control
| Pattern | Level | Evidence |
|---------|-------|----------|
| Command Palette | [emoji] | [What was found] |
| Keyboard Shortcuts | [emoji] | [What was found] |
| Focus Management | [emoji] | [What was found] |
### Editing & Data Entry
| Pattern | Level | Evidence |
|---------|-------|----------|
| Inline Editing | [emoji] | [What was found] |
| Strong Defaults | [emoji] | [What was found] |
| Form Efficiency | [emoji] | [What was found] |
### Navigation & Context
| Pattern | Level | Evidence |
|---------|-------|----------|
| Minimal States | [emoji] | [What was found] |
| Search-First | [emoji] | [What was found] |
| List-Centric | [emoji] | [What was found] |
### Feedback & Performance
| Pattern | Level | Evidence |
|---------|-------|----------|
| Optimistic UI | [emoji] | [What was found] |
| Response Time | [emoji] | [What was found] |
| Undo/Redo | [emoji] | [What was found] |
### Visual Design
| Pattern | Level | Evidence |
|---------|-------|----------|
| Visual Clarity | [emoji] | [What was found] |
| Context > Settings | [emoji] | [What was found] |
| Information Density | [emoji] | [What was found] |
---
## 🔴 Critical Issues
> Blocks keyboard users or forces unnecessary navigation
### [Issue Title]
**Pattern violated:** [Which efficiency pattern]
**Location:** `path/to/component.tsx`
[What the user experiences and why it slows them down]
**Fix:** [Specific recommendation]
---
## 🟠 Friction Points
> Slows down power users
- **[Location]** — [What's slow and why]
- **[Location]** — [What's slow and why]
---
## 🟡 Polish Opportunities
> Would elevate to pro-tool level
- [Opportunity 1]
- [Opportunity 2]
---
## ✅ What's Already Great
- [Positive finding 1]
- [Positive finding 2]
---
## Edge Cases
| Scenario | Status | Notes |
|----------|--------|-------|
| Empty state | [emoji] | [Has CTA? Keyboard accessible?] |
| Error state | [emoji] | [Dismissible? Undo available?] |
| Loading state | [emoji] | [Skeleton? Optimistic?] |
| Bulk operations | [emoji] | [Keyboard select? Batch actions?] |
---
## Quick Wins
> Low effort, high impact improvements
1. [ ] [Quick win 1]
2. [ ] [Quick win 2]
3. [ ] [Quick win 3]
---
## Summary
| Category | Score |
|----------|-------|
| Command & Control | [X/3 patterns at Good+] |
| Editing & Data Entry | [X/3 patterns at Good+] |
| Navigation & Context | [X/3 patterns at Good+] |
| Feedback & Performance | [X/3 patterns at Good+] |
| Visual Design | [X/3 patterns at Good+] |
| **Overall** | [X/15 patterns at Good+] |
Step 6: Offer Next Steps
After presenting the report, offer actions:
- Add keyboard shortcuts — help implement missing shortcuts
- Convert to inline editing — refactor modals to inline patterns
- Add to command palette — register new actions in ⌘K
- Implement optimistic UI — add optimistic updates for slow operations
- Re-review — run the review again after changes
Severity Levels
| Level | Criteria | Examples | |-------|----------|----------| | 🔴 Critical | Keyboard users blocked; forced full-page navigation for simple actions | No keyboard access to primary action; modal required for single-field edit | | 🟠 Friction | Power users slowed down; inconsistent with app patterns | Missing shortcut for frequent action; spinner where optimistic would work | | 🟡 Polish | Good but not pro-level; minor efficiency gains possible | Could add to command palette; could show undo toast |
Arguments
$ARGUMENTS- Spec name, file paths, or review scope<spec-name>— review UX based on spec documents and related code<file-path>— review UX of specific files or directoriesbranch— review UX of all changes on current branch vs main
Examples:
review:ux user-auth— review UX for the user-auth feature specreview:ux src/pages/settings/— review UX of the settings pagesreview:ux branch— review UX of all UI changes on the current branch
Du fichier au premier essai
Le téléchargement contient exactement le contenu affiché sur cette fiche. Cette copie peut différer de la dernière version du dépôt source.
Consulter la source et ses fichiers annexesCe téléchargement contient uniquement SKILL.md. Les scripts, références, modèles et dépendances éventuels ne sont pas inclus. Récupérez les fichiers nécessaires depuis la source en conservant leur arborescence avant de poursuivre.
1. Télécharger le fichier
Télécharger SKILL.mdDepuis la racine d’un projet de test, enregistrez le fichier sous .skill-downloads/ux-review-mrzxuqb9/SKILL.md, ou utilisez la commande suivante. Ce dossier sert à examiner la skill avant son installation.
Commande pour macOS/Linux ou WSL : un shell POSIX et curl sont nécessaires. Sous Windows sans WSL, utilisez le bouton de téléchargement et créez le dossier manuellement.
mkdir -p '.skill-downloads/ux-review-mrzxuqb9' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/ux-review-mrzxuqb9/download' --output '.skill-downloads/ux-review-mrzxuqb9/SKILL.md.part' && mv '.skill-downloads/ux-review-mrzxuqb9/SKILL.md.part' '.skill-downloads/ux-review-mrzxuqb9/SKILL.md'2. Vérifier les prérequis
Lisez SKILL.md avant de l’utiliser. Si le fichier cite des scripts ou services externes, examinez-les et préparez les accès nécessaires dans votre environnement de test.
Aucune section de prérequis reconnue dans ce fichier. Cela ne signifie pas qu’il est autonome : les bibliothèques, outils et accès peuvent être cités ailleurs dans le contenu ou dans le dépôt source.
Les plateformes de la fiche sont des indications du catalogue. Les chemins ci-dessous suivent la documentation des outils ; ils ne prouvent pas que cette skill y a été exécutée. Aucune version minimale de ces outils n’est attestée pour cette fiche.
3. Charger la skill
Le fichier ne fournit pas un en-tête name/description exploitable pour cette procédure native. Le téléchargement reste disponible pour lecture ; utilisez le mode manuel ci-dessous.
Mode manuel ou autre outil : joignez SKILL.md à une conversation, ou ouvrez-le dans un agent capable de lire les fichiers. Ajoutez les ressources annexes nécessaires. Il n’existe pas de dossier universel reconnu par tous les outils.
4. Faire un premier essai
Utilisez une entrée courte et fictive : un endpoint d’API, un petit document ou un brief d’interface. Dans une nouvelle conversation, joignez le fichier ou donnez son chemin à l’agent, puis adaptez cette demande :
Lis .skill-downloads/ux-review-mrzxuqb9/SKILL.md et applique ses instructions à cet exemple : [décrire une entrée de test]. Indique les fichiers, dépendances ou accès qui manquent. Produis le résultat demandé par la skill sans inventer de données ni exécuter de script non examiné.Résultat attendu : une sortie correspondant aux instructions de la skill, ou une liste précise des prérequis manquants. Une simple confirmation de lecture ne valide pas le premier usage. Si une ressource externe est indisponible, ce parcours reste à tester.
Bonnes pratiques React Hook Form
Developpement
Maîtrisez React Hook Form : évitez les anti-patterns, gérez les abonnements, et traitez correctement reset, dirty state et entrées contrôlées.
Migration Emotion vers Mantine
Developpement
Migrer les styled-components Emotion vers des composants Mantine avec style props et modules CSS. Pour convertir .styled.tsx ou supprimer les imports @emotion.
Détecteur de dérive de périmètre
Developpement
Analyse les diffs git par rapport à une intention pour détecter la dérive de périmètre, fichiers non liés et PR trop larges, puis suggère de garder, diviser ou justifier.