Design & Media
Design agent skills guide AI agents through visual and UX work: UI mockups, design systems, image and asset workflows, and design reviews. A skill keeps the agent aligned with your design language instead of improvising a new style each session.
-
qredence Skill Library Health ReportAudits accessible components, styles, and variables for naming, duplication, documentation, and structural health. Use for a scoped library review; include usage or publication findings only when the required library or analytics context is explicitly available.
-
qredence Skill Motion Spec GeneratorDefines and documents a precise motion spec (duration, easing curve, delay, properties animated) for key transitions in a flow, so the intended motion feel is captured explicitly rather than left to be reverse-engineered from a prototype. Use when documenting motion decisions for engineering handoff or design system consistency.
-
qredence Skill Site Launch ChecklistReviews a Figma Sites project against a pre-launch checklist: responsive behavior across breakpoints, working links/navigation, page titles and meta content, image alt text, and load-affecting asset sizes. Use before publishing a Figma Sites project.
-
qredence Skill Localization ReadinessChecks a design for issues that will surface once text is translated: fixed-width containers that will clip longer languages, RTL layout assumptions, concatenated/hardcoded sentence fragments, and untranslatable content baked into images. Use before sending a design or file for localization.
-
qredence Skill Shadcn Theme VariablesAdd, repair, or normalize Figma Variables so a Figma design file matches the shadcn/ui semantic theming model. Use when the user asks to create shadcn theme variables, fix light/dark theme variables, map Figma Variables to shadcn CSS tokens, align variables with app/globals.css, or update variable-bound properties for shadcn-style components.
-
qredence Skill Spacing Scale EnforcerMaps arbitrary spacing values to the nearest step on the project's spacing scale (typically Tailwind's 4px-based scale) and flags off-scale spacing in designs or code. Use when spacing looks inconsistent or when translating Figma spacing values into Tailwind utility classes.
-
qredence Skill Figma To Code ComponentTranslates a selected Figma frame or component into idiomatic code for a specified target (e.g. React + Tailwind, React + CSS Modules, SwiftUI, Jetpack Compose), mapping Figma variables to design tokens and Figma auto layout to the target's layout system. Use for a first-pass implementation of a specific design element, not a full application build.
-
qredence Skill Screenshot To ComponentUse when the user provides a screenshot or crop of a UI component, panel, sidebar, modal, menu, card, toolbar, form, or dense app surface and wants it recreated as a structured Figma Design component. Focuses on editable hierarchy, auto layout, semantic nesting, reusable subcomponents, component properties, variants, and maintainable layer naming instead of flat screenshot tracing.
-
qredence Skill Visual Consistency CheckScans a flow for inconsistent spacing, alignment, corner radii, icon styles, and grid usage across screens that should feel like one cohesive product. Use when multiple screens were designed at different times or by different people and need to feel unified.
-
qredence Skill Design First UI PromptingTurn a vague product, page, screen, component, or prototype request into a Figma-ready prompt with clear goal, audience, layout, typography, color, content, states, constraints, and non-goals. Use before asking Figma Agent or Figma Make to generate, revise, or refine UI when the user needs a stronger prompt rather than direct design execution.
-
qredence Skill Plugin Widget RecommenderRecommends plugin or widget categories for a concrete Figma workflow and audits supplied plugin usage against a supplied approval policy. Use when the user needs tool-selection guidance or a policy comparison; verify current marketplace availability and do not assume access to file-level plugin history or organization admin data.
-
qredence Skill States Completeness CheckChecks that every screen and key component in a flow has the full set of states it needs -- empty, loading, populated, error, partial/edge-case data, and (for components) hover/focus/pressed/disabled -- and lists which are missing. Use before a flow is considered complete for handoff.
-
qredence Skill Token Tailwind Theme SyncSyncs Figma variables (color, spacing, radius, typography) into a tailwind.config theme extension and CSS custom properties, following shadcn's theming file structure. Use when Figma variables have changed and the Tailwind/CSS theme needs to catch up, specifically for shadcn-based projects.
-
qredence Skill Variable Driven PrototypeDesigns a state model for a variable-driven Figma prototype, including variables, initial values, set-variable moments, conditional branches, reflected UI states, and manual verification steps. Use when a prototype needs remembered state or branching logic, but treat the result as a specification unless the current Figma Agent session explicitly supports prototype interaction editing.
-
qredence Skill Legacy Styles To VariablesFinds hardcoded fill, stroke, and effect colors, hardcoded spacing values, or old-style color/text styles in a file and migrates them to the equivalent shared variables or styles from an enabled library. Use when adopting variables in a file that predates them, or migrating a file onto a newer design system version.
-
qredence Skill Naming Convention EnforcerRenames layers, frames, components, and pages to follow a specified naming convention (e.g. PascalCase components, sentence-case screens, prefix-based icon names). Use for cleaning up messy layer names before handoff, publishing, or Dev Mode inspection.
-
qredence Skill Shadcn Component StructureEnsures generated React components follow shadcn/ui structural conventions -- forwardRef/ref handling, cn() className merging, cva-based variants, proper prop interfaces extending native element types, and correct file placement. Use when generating or reviewing component code meant to fit into a shadcn/ui-based codebase.
-
qredence Skill Tailwind Class Order CheckAudits className strings for consistent ordering, duplicate or conflicting utility classes, and alignment with the project's class-sorting convention. Use when reviewing generated or hand-written component code for Tailwind class hygiene.
-
qredence Skill Animation Consistency CheckAudits animated components and prototype transitions across a file for consistent easing, duration, and motion character, flagging transitions that feel out of step with the rest of the product's motion language. Use when multiple people or sessions have added animation and it needs to feel cohesive.
-
qredence Skill Container Layout NormalizerSets consistent container widths, max-width, centering, and responsive padding across breakpoints, matching standard Tailwind/shadcn container conventions instead of ad hoc per-page values. Use when a layout's outer container sizing looks inconsistent or arbitrary.
-
qredence Skill Responsive Breakpoint CheckReviews a layout's auto layout, constraints, and resizing settings to verify it will adapt correctly across specified breakpoints (e.g. mobile, tablet, desktop) rather than just checking how it looks at one frame size. Use when a design needs to work responsively, especially for Figma Sites or handoff to a responsive web build.
-
qredence Skill Base UI Primitive CompositionComposes interactive components (dialogs, popovers, tooltips, menus, tabs, accordions) from Base UI's unstyled primitive parts (Root/Trigger/Portal/Content/etc.) instead of hand-rolled behavior, preserving built-in accessibility and state management. Use when building any interactive/overlay component that has a Base UI or Radix equivalent primitive.
-
qredence Skill Color Token Format NormalizerConverts colors to the CSS-variable/HSL token format used by shadcn's theming convention (--background, --primary, etc.), flags hardcoded hex/rgb values that should reference a token, and keeps light/dark pairs aligned. Use when generating or auditing color usage in a shadcn-based codebase.
-
sylphxai Bundle UI FlowDesign or implement one product UI flow with loading, empty, error, and success states, verified on a real render. Use when the user asks to build a screen, fix a form, design a flow, or match an existing design system. Do not use to review an existing flow, extract a design system, or produce store screenshots.
-
carlheath Bundle DesignUX/UI design specialist for user experience, interface design, and visual communication. [VAD] UI design, UX research, wireframes, prototypes, user flows. Design systems, component libraries, accessibility (WCAG). [NÄR] Use when: design, UI, UX, interface, wireframe, prototype, mockup, user flow, design system, visual design, accessibility [EXPERTISE] User-centered design, visual design, UI patterns, accessibility
-
carlheath Bundle ArchitectureSystem architecture and technical design specialist. [VAD] System design, architecture patterns, ADRs, technology evaluation. Cloud/microservices architecture, integration patterns, API design. [NÄR] Use when: architecture, design system, system design, ADR, tech stack, scalability, arkitektur, systemdesign, teknikval, skalbarhet [EXPERTISE] Architecture patterns, trade-off analysis, scalability, integration
-
carlheath Skill Image CreatorImage generation and image editing. Two engines: GPT-Image-1 (OpenAI) and Gemini Nano Banana Pro (Google). Seven prioritised workflows: blog cover, presentation illustration, concept sketch, geopolitical map, meme/satirical image, infographic prompt, image edit/inpainting. [WHAT] Image skill via `~/.claude/scripts/call-image-api`. Default: Nano Banana Pro for photorealism and maps, GPT-Image-1 for precision edits and text-in-image. Output saves to a configured pictures path. [WHEN] Use when: generate image, illustration, AI image, cover image, presentation image, infographic, image edit/inpainting.
-
danielleit241 Bundle Hs UI UX Pro MaxUI/UX design intelligence for web and mobile: style selection, color systems, typography, layout, accessibility, interaction states, responsive behavior, forms, charts, design systems, and code review across React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, and HTML/CSS.
-
brandonburrus Bundle Edit ImageThis skill should be used when manipulating or converting raster images with sharp, including "convert this png to jpg", "resize this image", "make a thumbnail", "compress this photo", "crop/rotate/flip this image", "strip EXIF metadata", "add a watermark", "overlay my logo", "batch convert these images", or "what are this image's dimensions". Covers jpeg, png, webp, avif, gif, tiff. It should not be used for creating diagrams (use create-diagram), charts or plots from data (use visualize-data or analyze-data), editing PDFs (use edit-pdf), or vector/SVG authoring.
-
jgamaraalv Bundle Figma Code ConnectCreate or update Figma Code Connect template files (.figma.ts/.figma.js) that map Figma components to code. Use for Code Connect, Figma component mapping, design-to-code.
-
willakins Bundle Create PDFCreate or revise polished PDF deliverables with intentional layout, typography, and deterministic source files. Use when Codex needs to build a PDF from scratch or update an existing styled PDF, especially for business documents such as project outlines, proposals, reports, summaries, implementation plans, or other static documents where preserving page size, hierarchy, tables, banners, footers, and pagination matters.
-
kinhluan Bundle Conference PaperWrite conference papers (4-6 pages), poster abstracts, and prepare oral/poster presentations. Use when targeting conferences like NeurIPS, ICML, CVPR, ACL, IEEE venues, or preparing conference presentations.
-
robsonrung Bundle UI UX Pro MaxChoose a reusable UI design system or review cross-screen UX using a design database. Use for visual systems, interaction rules, or UI quality; use frontend-design for bespoke visual implementation.
-
trac3r00 Bundle DesignClaude-grade visual design outside Claude Code — distinctive web/UI design direction, poster and canvas art philosophy, and a 10-theme styling library, portable to any agent that reads SKILL.md. Use when building or restyling a UI, landing page, poster, slide deck, static art piece, or any artifact that must not look AI-generated.
-
sandeco Skill Mira Image TemplateCria um NOVO template de deck do Mira a partir de imagem(ns) de referência (prints de slides e/ou logomarca): reconhece o design system e a disposição dos elementos e salva o esqueleto index.html (design system no bloco @MIRA:THEME) mais o tema em mira-templates/, para aparecer no /mira-new. Use SEMPRE que o usuário disser "/mira-image-template", "criar template a partir de imagem", "template a partir desse print", "transforma essa tela em template", "monta um template com essa identidade", "usa essa logomarca como template", "extrai o design system dessa imagem", "novo template de deck a partir de imagem", ou anexar imagens pedindo para virarem um template/identidade de slides.
-
sandeco Bundle Mira Story ArchitectArquitetar histórias completas e de alto impacto para o MIRA a partir de uma premissa ou contrato conceitual, gerando premissa validada, sete passos estruturais, rede de personagens, tema ou argumento moral, mundo ficcional, rede de símbolos, trama, teia de cenas, construção de cenas e diálogo sinfônico. Usar quando o usuário pedir para desenvolver, estruturar, aprofundar ou auditar uma história narrativa para slides, animação ou roteiro MIRA. Não usar para pesquisar notícias e gerar apenas premissas, nem para montar ou animar o deck, que são do /mira-builder e do /mira- animator.
Frequently asked questions
What are Design & Media agent skills?
Design agent skills guide AI agents through visual and UX work: UI mockups, design systems, image and asset workflows, and design reviews. A skill keeps the agent aligned with your design language instead of improvising a new style each session.
Which Design & Media skills are most installed?
Popular Design & Media skills on SkillMD right now include library-health-report, motion-spec-generator, site-launch-checklist. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do Design & Media skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.