app-design

À utiliser lors de la construction ou de la modification de la mise en page de l'application, de la création de composants d'interface utilisateur ou de toute décision de conception visuelle. Garantit la cohérence, l'accessibilité et une...

npx skills add https://github.com/microsoft/fabric-apps-analytic-templates --skill app-design

App Design

Your job is to build cohesive, distinctive apps — not just correct ones. Add character through intentional design decisions — typography pairing, color emphasis, spatial rhythm, and a clear visual point of view.

Aesthetic Direction

Before building anything, decide the app's tone in one word and its signature detail — the one thing someone notices first. These two choices guide every decision that follows. Pick a tone that is specific and bold, not safe or generic. Examples like editorial, geometric, organic, industrial, playful, or minimal are starting points — don't limit yourself to these. Invent a direction that fits the app's purpose.

Then match your execution to your direction — a maximalist direction needs layered effects and rich detail in the code; a minimalist direction needs precise spacing, restraint, and careful typography. Elegance comes from committing to the direction fully, not from adding more.

Typography

Pick fonts that set the app's character — this is one of the strongest signals of intentional design. At minimum choose a characterful --font-heading paired with a complementary --font-base, ideally from the same foundry or design family. Update --font-monospace and --font-numeric if necessary. Avoid generic defaults like Arial, Inter, or Roboto.

Load fonts via Google Fonts (or another CDN) as <link> tags in index.html, then update the font family tokens in the @theme block of global.css.

Theming Workflow

Start by customizing src/global.css — this is the single source of truth for the app's active visual identity. Every component uses these tokens, so setting them first means the entire UI shifts together.

  1. Colors: Update the semantic color tokens (--color-primary, --color-background, --color-card, --color-border, etc.) in both the @theme block and the .dark override to match the aesthetic direction. The defaults are neutral blue/grey — make them yours.
  2. Data palette: Define the ten data color tokens (--color-data-1–--color-data-10) as the active palette, with .dark overrides, per the data color rules. Then populate src/data-palette-presets.json with three alternative palettes — never a copy of the active one — each with a kebab-case id, a concise name, and colors.light / colors.dark arrays of exactly ten six-digit hex colors in token order.
  3. Radius: Adjust --radius (the base radius) and the radius scale to match the tone — sharp/geometric (lower values), soft/rounded (higher values), or pill-shaped (--radius-full).
  4. Fonts: Update the font family tokens as described in the Typography section above.
  5. Then build components. Focus component-level styling on layout, spacing, and element-specific details — not re-specifying colors and radii that the tokens already handle.
  6. Selective overrides last. After the base theme is in place, inspect and adjust individual components that need to deviate — an accent-colored card border, a button with a unique hover effect, etc.

Keep these principles in mind:

  • Use semantic color tokens so surfaces, text, and borders adapt correctly to light and dark mode.
  • Maintain sufficient contrast between text and backgrounds.
  • The recommended minimum text size is text-200.
  • Keep spacing consistent — use the spacing tokens from the theme rather than arbitrary values.
  • Ensure all interactive elements are keyboard-accessible with visible focus indicators and appropriate disabled states.

Read these reference files — they include "Make it yours" prompts that tie back to the aesthetic direction above:

  • UI Style Recipes — per-element styling guidance for buttons, cards, inputs, dialogs, tabs, tooltips, tables, and more.
  • Visual Style Recipes — chart theming, Vega-Lite config, dark mode chart support, and mark-specific styling.

App Layout

These are good defaults for app structure. Adapt them to the specific app's needs.

Page Structure

  • The app layout should fill the viewport.
  • On wide screens, constrain the content width so it doesn't stretch uncomfortably. Use responsive breakpoints or multi-column layouts to make good use of available space.

Don't default to the same layout every time. The structure should serve the aesthetic direction — a sidebar + main content split, a full-width single column, a multi-panel master-detail, an asymmetric split, or something else entirely. These are starting points, not an exhaustive list. Invent a layout that fits the app's purpose and tone.

The header/toolbar is part of the design language — not every app needs a traditional fixed header. Consider alternatives: a floating command bar, a compact inline toolbar, a branded banner, a collapsible drawer, a minimal top-right action cluster, or no header at all if the content speaks for itself.

Container Sizing

VegaVisual and other content components fill their container — the container controls their dimensions. Without constraints, charts and content stretch to the full viewport width, which produces squished, unreadable visuals on wide screens.

  • Constrain the dashboard wrapper, not individual charts. Apply a max-width to the outermost content wrapper that holds the dashboard. This single constraint keeps the entire layout proportional on wide monitors. If the app lacks an outer wrapper, create one.
  • Do not constrain individual chart containers. Let the wrapper width plus grid columns determine each chart's size naturally.
  • If the user explicitly requests full-width charts, confirm the design choice before proceeding.

Dashboard Grid

  • Start mobile-first and scale up columns with responsive breakpoints.
  • Support mixed-size cards via span utilities.

Avoid uniform grids where every card is the same size — they look like a spreadsheet. Vary card spans to create visual hierarchy: a wide chart spanning two columns next to a tall narrow KPI panel, or a full-width table below a row of smaller cards. Let the data importance guide which elements get more space.

Loading, Empty & Error States

Every component that depends on async data should handle all three states:

StateWhat to show
LoadingA skeleton placeholder matching the shape of the expected content
EmptyA centered muted message explaining no data is available
ErrorA destructive-styled banner with the error message

Dark Mode

Include a light/dark mode toggle in the app header or toolbar. Use the useAppTheme hook from @/hooks/use-theme to read and toggle the theme.


Coding Conventions

  • Styling: Tailwind CSS v4 utility classes for all styling. Theme colors are defined as CSS custom properties in src/global.css using @theme. Use Tailwind classes directly in JSX.
  • Theming: Light/dark color tokens defined in src/global.css via CSS custom properties. Dark mode uses the .dark class on the root html element, auto-detected via prefers-color-scheme, data-appearance attribute, or .dark class. The useAppTheme hook in src/hooks/use-theme.ts manages the toggle.
  • CSS class merging: Use cn() from @/lib/utils (powered by clsx + tailwind-merge) to conditionally combine Tailwind class names.
  • Icons: Lucide React for UI icons.
  • UI Components: Use Radix primitives with Tailwind CSS styling for all interactive elements — buttons, inputs, dialogs, menus, tabs, etc.

UI Token Rules

All styling must use the design tokens defined in src/global.css via Tailwind utility classes. Never hardcode raw color values, pixel sizes, or font stacks — raw values are only permitted in global.css and src/data-palette-presets.json. Refer to global.css for available tokens, their values, and expected usage.

Examples:

  • bg-primary text-primary-foreground — not bg-blue-600 text-white
  • text-300 — not text-sm or text-[14px]
  • p-400 gap-300 — not p-4 or gap-3
  • font-semibold — not font-[600]
  • rounded-xl — not rounded-[8px]
  • icon-size-200 — not w-4 h-4

cn() and tailwind-merge conflicts: tailwind-merge treats text-* utilities as one conflict group. In cn(), combining text size and text color with ambiguous text-* classes can drop one class. Prefer explicit length syntax for font size (e.g., text-[length:var(--text-300)]) when combining with text color classes inside cn(). If classes are static and not merged, text-300 text-foreground is acceptable.

Form element font inheritance: Native form controls may not inherit the page font family by default. Ensure base styles in global.css set font-family: inherit for select, input, textarea, and button.


Final Audit

After assembling a layout, audit each element in its actual context — not in isolation. A component may look correct on its own but break the visual rhythm of the page. Check: Is every text legible? Are labels proportional to their controls? Are toolbar rows aligned on a shared edge? Do charts fill their containers? Do repeated elements (badges in tables, icons in lists) maintain appropriate visual weight for their density? Fix anything that fails.

Plus de skills de microsoft

oss-growth
microsoft
Persona de growth hacker OSS
agent-framework-azure-ai-py
microsoft
Créez des agents Azure AI Foundry à l’aide du SDK Python Microsoft Agent Framework (agent-framework-azure-ai). À utiliser lors de la création d’agents persistants avec AzureAIAgentsProvider, de l’utilisation d’outils hébergés (interpréteur de code, recherche de fichiers, recherche web), de l’intégration de serveurs MCP, de la gestion de fils de conversation ou de l’implémentation de réponses en streaming. Couvre les outils de fonction, les sorties structurées et les agents multi-outils.
development
airunway-aks-setup
microsoft
Configurez AI Runway sur AKS — du cluster nu au modèle en cours d'exécution. Couvre la vérification du cluster, l'installation du contrôleur, l'évaluation GPU, la configuration du fournisseur et le premier déploiement. QUAND : « configurer AI Runway », « intégrer un cluster AKS », « installer AI Runway », « configuration airunway », « déployer un modèle sur AKS », « inférence GPU sur AKS », « configuration KAITO sur AKS », « exécuter LLM sur AKS », « vLLM sur AKS », « configurer le service de modèles sur AKS », « contrôleur AI Runway ».
devops
appinsights-instrumentation
microsoft
Conseils pour instrumenter les applications web avec Azure Application Insights. Fournit des modèles de télémétrie, la configuration du SDK et des références de configuration. QUAND : comment instrumenter une application, SDK App Insights, modèles de télémétrie, qu'est-ce qu'App Insights, conseils sur Application Insights, exemples d'instrumentation, bonnes pratiques APM.
devops
applicationinsights-web-ts
microsoft
Instrumentez les applications navigateur/web avec le SDK JavaScript Application Insights (@microsoft/applicationinsights-web). Utilisez-le pour la surveillance des utilisateurs réels (RUM) — vues de page, clics, dépendances AJAX/fetch, exceptions, événements personnalisés et traces d’agents GenAI côté navigateur corrélées aux traces OpenTelemetry backend. Couvre le script de chargement du SDK et la configuration npm, les extensions de framework (React, React Native, Angular), Click Analytics, les initialiseurs de télémétrie et les conventions sémantiques OTel GenAI pour les spans d’agents/outils/modèles émises depuis le navigateur.
devops
azure-ai-anomalydetector-java
microsoft
Créez des applications de détection d'anomalies avec le SDK Azure AI Anomaly Detector pour Java. Utilisez-le lors de l'implémentation de la détection d'anomalies univariées/multivariées, de l'analyse de séries temporelles ou de la surveillance basée sur l'IA.
development
azure-ai-language-conversations-py
microsoft
Implémentez la compréhension du langage conversationnel (CLU) à l’aide du SDK Python azure-ai-language-conversations. Utilisez-le lorsque vous travaillez avec ConversationAnalysisClient pour analyser l’intention et les entités d’une conversation, créer des fonctionnalités de NLP ou intégrer la compréhension du langage dans des applications.
development
azure-ai-ml-py
microsoft
SDK v2 d’Azure Machine Learning pour Python. Utiliser pour les espaces de travail ML, les tâches, les modèles, les jeux de données, le calcul et les pipelines. Déclencheurs : « azure-ai-ml », « MLClient », « espace de travail », « registre de modèles », « tâches d’entraînement », « jeux de données ».
development