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.
-
majiayu000 Bundle Paper Illustration Image2Generate publication-quality academic illustrations through a local Codex app-server bridge that uses Codex native image generation. This is a separate experimental alternative to `paper-illustration`, intended for Claude Code users who want a GPT-image-style renderer without modifying the original skill.
567 -
majiayu000 Bundle Lp Do Assessment 14 Logo BriefLogo design brief for new startups (ASSESSMENT-14). Synthesises brand identity dossier, brand profile, and product naming to produce a complete logo design brief ready to hand to a designer. Upstream of lp-do-assessment-15-packaging-brief and /lp-design-spec.
567 -
majiayu000 Bundle Stitch MCP Apply Design SystemApplies a Stitch Design System to existing screens — updates their colors, fonts, and roundness to match the design system's theme. Requires an assetId from list or create operations.
567 -
majiayu000 Bundle Svelte Game Visual DesignGraphical and UI/UX design improvements for SvelteKit browser games, including layout, typography, color systems, HUD design, visual polish, and dynamic effects. Use when asked to redesign or enhance a Svelte game's visuals, to improve readability or aesthetics, to add motion/animation polish, or to audit/iterate on UI using Playwright MCP snapshots.
567 -
majiayu000 Bundle Remix Graphic Ad From ReferenceRecreate a static graphic ad (Pinterest pin, IG/FB feed image, poster) from a reference image, swapping in a new brand's product and new copy while keeping the reference's layout, composition, and visual energy. ALWAYS generated with GPT Image 2 in edit-the-reference mode (fal-ai/gpt-image-1/edit-image, a billed FAL generation); the HTML/goose-graphics overlay is only an optional text-finishing step, never the generator. The static-graphics counterpart to the video remix-ad skill; this is what the app calls when a user picks a reference ad and wants it for their own product.
567 -
majiayu000 Bundle Color Theory Palette Harmony ExpertExpert in color theory, palette harmony, and perceptual color science for computational photo composition. Specializes in earth-mover distance optimization, warm/cool alternation, diversity-aware palette selection, and hue-based photo sequencing. Activate on "color palette", "color harmony", "warm cool", "earth mover distance", "Wasserstein", "LAB space", "hue sorted", "palette matching". NOT for basic RGB manipulation (use standard image processing), single-photo color grading (use native-app-designer), UI color schemes (use vaporwave-glassomorphic-ui-designer), or color blindness simulation (accessibility specialists).
567 -
majiayu000 Bundle Figma Atomic Design ImplementationImplement or refactor Figma screens in Angular using strict Atomic Design order (atoms -> molecules -> organisms -> templates -> pages), reusable component APIs, and complete companion files (.ts, .html, style, .stories.ts, .spec.ts). Use for Figma-to-code delivery, design-system normalization, component decomposition, and atomic architecture cleanup.
567 -
majiayu000 Bundle Lp Do Assessment 11 Brand IdentityVisual brand identity for new startups (ASSESSMENT-11). Reads brand-strategy artifact and produces brand-dossier with colour palette, typography, imagery direction, and token overrides. Upstream of S1.
567 -
majiayu000 Bundle Render Vo Anchored Motion ListicleAssemble an expert/educator motion-graphic LISTICLE video ad from a config — a spoken authoritative voiceover carries a numbered listicle while N web-animated hyperframe beats (HTML plus the Web Animations API, one branded design system of alternating tiles, big hero numerals, and glass-pill callouts) are rendered frame-by-frame via Playwright and anchored to the VO's word-level timestamps, periodic color-graded B-roll windows give visual breath, and captions burn ONLY inside those B-roll windows (2-word chunks, ASS Format header carrying a Name field so none drop) with the VO mixed under a low music bed. This is the FREE deterministic assembly stage (Playwright beat render plus ffmpeg concat plus window-masked caption burn plus VO-and-music mix plus final composite) — the VO, the music bed, and the stock B-roll come from create-vo-elevenlabs, create-music-elevenlabs, and media-proxy. Use for the vo-anchored-motion-listicle format.
567 -
majiayu000 Bundle Flutter Bloc DevelopmentBuild Flutter features using BLoC state management, clean architecture layers, and the project's design system. Apply when creating screens, widgets, or data integrations.
567 -
majiayu000 Bundle Lp Do Assessment 15 Packaging BriefPackaging design brief for new startups (ASSESSMENT-15). Conditional on physical-product profile. Reads brand identity, product naming, logo brief, and distribution plan to produce a complete packaging brief covering structural format, regulatory requirements, surface design, and designer handoff checklist.
567 -
majiayu000 Bundle Ptsf Sponsor Logo UploadUse when uploading sponsor logos to ProductTank SF Google Drive - downloads image from URL (handles Cloudflare), validates it's actual image data, uploads to Drive folder, returns shareable link
567 -
majiayu000 Bundle Design Systems 3Build, maintain, and scale design systems including component libraries, design tokens, documentation, and governance models. Use for creating design systems from scratch, establishing component architecture, implementing design tokens for multi-brand/multi-theme support, building scalable component libraries, setting up design-to-code workflows, creating design system documentation, establishing governance processes, managing design system teams, and ensuring consistency across products and platforms.
567 -
majiayu000 Bundle Frontend Design 4Design thinking and decision-making for web UI. Use when designing components, layouts, color schemes, typography, or creating aesthetic interfaces. Teaches principles, not fixed values.
567 -
majiayu000 Bundle Frontend Excellence 2Create distinctive, production-grade frontend interfaces that avoid generic AI aesthetics. Use when building web UIs, React components, landing pages, dashboards, or styling any web interface. Provides design thinking process, color palettes, typography treatments, and anti-patterns to avoid.
567 -
majiayu000 Bundle Prototype 5Advanced prototyping (Artifacts/Figma/Lovable/v0/Bolt)
567 -
majiayu000 Bundle UI Designer 3Highly opinionated user interface designer. Turns vague screens into concrete interface direction with strong hierarchy, spacing, typography, states, and component rules. Hates decorative noise, weak affordances, inconsistent components, and layouts that make users hunt for the next action. Use for dashboards, forms, tables, mobile screens, landing pages, redesigns, design-system decisions, visual hierarchy critiques, and any request about how an interface should look and behave.
567 -
majiayu000 Bundle UI UX Design 3UI/UX design systems, Figma workflows, accessibility checklists, browser-based visual audit with automated screenshots, component specification patterns, and design-to-code handoff procedures. Use when designing interfaces, creating design systems, auditing accessibility, reviewing existing UI via browser screenshots, specifying components, or preparing design handoff for developers.
567 -
majiayu000 Bundle Component Design 2Design reusable UI components and design system elements with consistent styling and behavior. Use for: building design systems, creating component libraries, establishing UI patterns, and ensuring design consistency across products.
567 -
majiayu000 Bundle Design Guide 4CoHalo UI design system guide for building consistent, reusable frontend components. Use when creating new UI components, modifying existing ones, adding pages or features to the frontend, styling UI elements, or when you need to understand the design language and conventions. Covers: component creation, design tokens, typography, status/priority systems, composition patterns, and the /design-guide showcase page. Always use this skill alongside the frontend-design skill (for visual quality) and the web-design-guidelines skill (for web best practices).
567 -
majiayu000 Bundle Design Spec 3[Project Management] Create UI/UX design specifications from requirements, PBIs, or user stories. Produces structured design spec documents with layout, typography, colors, interactions, and responsive breakpoints. Triggers on design spec, design specification, UI specification, component spec, layout spec, wireframe, mockup.
567 -
majiayu000 Bundle Design System 6Token architecture, component specifications, and slide generation. Three-layer tokens (primitive→semantic→component), CSS variables, spacing/typography scales, component specs, strategic slide creation. Use for design tokens, systematic design, brand-compliant presentations.
567 -
majiayu000 Bundle Frontend Design 7Design thinking and decision-making for web UI. Use when designing components, layouts, color schemes, typography, or creating aesthetic interfaces. Teaches principles, not fixed values.
567 -
majiayu000 Bundle Frontend UI 4Enterprise SaaS frontend — Tailwind v4, shadcn/ui, Next.js App Router or Vite SPA, accessibility, responsive design, component patterns. Use when: React component, Next.js page, frontend UI, Tailwind, shadcn, accessibility, a11y, responsive design, form validation, server component, client component, design system, dark mode, SaaS UI, dashboard, pricing page, enterprise UI, data table, landing page, Vite, React Router, SPA, single page app.
567 -
majiayu000 Bundle Reactive Md 2Create functional markdown documents with embedded interactive components for product team collaboration using the reactive-md VS Code extension. Use when generating product specs, design prototypes, feature wireframes, user flows, or interactive documentation. Specializes in teaching the 'live fence' methodology (css|jsx|tsx live) to replace Figma/Storybook workflows. NOT for production code—only for idea communication and rapid prototyping.
567 -
ahang1598 Skill Skill 27产品设计套件深挖工具。当用户是设计师要对 1-3 个竞品做深度 UX / 视觉 / 交互拆解,从中提炼"哪些值得借鉴、哪些必须回避、自己的差异化在哪"时启动的 Skill。与 PM 套件 `/竞品分析` 互补,服务不同视角——PM 套件做"市场地位 / Porter 五力 / SWOT / 定价 / 战略推演",本 Skill 做"信息架构 / 交互模式 / 视觉语言 / 内容策略 / 可借鉴细节",视角差异 ≠ 内容重复。 本 Skill 从设计师视角对竞品做三层识别(直接 / 间接 / 标杆)+ 3-7 个维度评估(IA / 交互模式 / 视觉 / 内容 / 性能 / 无障碍 / 移动端)+ Castle vs Shack 判断(哪些是结构性壁垒 vs 可复制表层)+ 不盲目复制的 3 问校验(为什么 work / 上下文一致吗 / 副作用是什么)+ 视觉策略决策(lean-in 对齐行业惯例 vs diverge 主动差异化)。核心输出是带标注截图引用的 `annotated_references[]` + 可执行的 `takeaways[]`(每条标 borrow / avoid / match 行动),下游 Audit 走查、Brief 策略、Pitch 提案都能直接消费。 触发关键词(前提:用户身份是设计师 / 设计师语境): - 设计师 + [竞品 / 友商 / 对标产品] [分析 / 拆解 / 解剖 / 学习] - 拆 [Linear / Notion / Figma 等具体产品] - 看看 [竞品 X] 是怎么做 [流程 / 页面 / 交互] 的 - 我们的 [功能 / 页面] 和 [竞品] 比差在哪 - 找设计借鉴 / 找视觉标杆 / 找交互模式参考 - 调研 [产品名] 的 [功能 / 模块 / 页面] - 帮我研究一下 [产品] 的 [功能 / 做法] - [产品名] 的 [功能] 是怎么做的 / 怎么实现的 - [产品名] + [功能] + 调研 / 研究 / 学习 - bench / competitor design teardown / UX competitive analysis ⚡ 路由消歧规则(当本 Skill 与 PM 套件 `/竞品分析` 冲突时): - 用户提到"调研/研究/拆/看看" + 单个具体产品名 + 具体功能/交互/页面 → 触发本 Skill(bench)
9 -
ahang1598 Bundle Skill 28视觉情绪板(Mood Board / Design System 起点)。基于 /用户旅程 + /用户故事 + /设计简报 上游 产出,推导 3-4 套色彩方向不同的设计方案,输出自包含 HTML(含氛围图 / 色板 / 字体 / 组件 demo 四行结构 + 标签页切换),用户选定后生成可被 /Web页面设计 /mobile页面设计 /动效规划 /数据可视化 消费的 design-tokens.json。 触发关键词:视觉情绪板、情绪板、视觉情绪、mood board、设计系统起点、风格推荐、color scheme、 设计方案推导、design tokens 起点、style definition、设计风格定义。 排除(反向):单屏视觉稿(用 /Web页面设计 或 /mobile页面设计)、动效规划(用 /动效规划)、 改版前体验诊断(用 /启发评估)、向上汇报材料(用 /设计提案)。
9 -
ahang1598 Skill Skill 41数字人头像生成专项。支持两种风格预设(realistic 写实摄影 / 3d-cartoon Pixar-Disney 3D 渲染)+ 两种主体(human 人 / non-human 动物・吉祥物・生物),用户在 Phase 1 显式确认后才进入生成。所有输出强制正脸 / 头不歪 / 直视镜头(永久锁定,不可协商);身体取景(headshot / half-body / full-body)+ 背景类型(solid / gradient / real-scene)+ 写实子风格(clean / polished / documentary)均按需用户确认,4 张图在确认参数上保持一致,仅在表情 / 光线 / 背景细节上做差异化。超出范围的风格(anime / 2D 插画 / 水墨 / 矢量 / 像素)会被识别并显式提示用户重映射或终止。 触发关键词:数字人、虚拟形象、品牌 IP 头像、客服头像、persona 头像、mascot、吉祥物、宠物头像、虚拟代言人、avatar、digital human、brand mascot。 排除(反向):UI / UX 页面设计(用 /Web页面设计 或 /mobile页面设计)、Logo / icon 设计(不在套件范围)、真人照片修图(不在 ImageGen 能力内)、动效 / 视频数字人(ImageGen 仅静态)、视觉情绪板(用 /board 当存在时)。
9 -
ahang1598 Skill Skill 43从营销主题生成可直接发布的完成品海报图片(PNG)。AI 直出材质化字体海报,文字融入画面材质 (霓虹/冰雕/木刻/水墨/玻璃/花卉/金箔/毛笔金 8 种材质风格),不用 Pillow 叠加。内置文字 正确性二次校验,错字自动重试(≤ 2 次)。支持多尺寸(16:9 公众号 / 3:4 小红书 / 1:1 朋友圈 / 9:16 故事)。可选消费 design-tokens.json 约束色彩。 触发关键词:营销图、头图、海报、banner、poster、封面图、hero image、公众号封面、小红书配图、 营销头图、活动海报、推广图、宣传图、社交媒体图。 排除(反向):UI / UX 页面设计(用 /Web页面设计 或 /mobile页面设计)、数据可视化图表(用 /数据可视化)、 数字人头像(用 /数字人头像)、视觉情绪板(用 /视觉情绪板)、Logo / icon 设计(不在套件范围)。
9 -
ahang1598 Bundle Skill 49设计系统提取(Design System Extract)。三模式:① 外部模式——薄包装 designlang CLI,从任意线上站点反向抽取完整 design tokens(颜色 / 字体 / 间距 / 圆角 / 阴影 / 动效等 19 类),产物供 /竞品拆解 / /视觉情绪板 / /设计简报 借鉴;② 内部模式——扫描自己的代码仓库(package.json / tailwind.config / components/ / 含 styled-components & emotion 的 CSS-in-JS AST 解析),把已实现的设计语言固化为 design.md;③ Figma 模式——通过 Figma MCP 从设计稿源头抽 Variables / Local Styles,作为设计师 native 入口。三种模式输出统一 schema(meta.source.type 区分),可链入下游 Brief / Board / Bench / 下一个项目 Flow Web,或作为 AI 编程(Cursor / Claude Code)的上下文锚点,也可反哺 SparkDesign 组件库。SparkDesign Diff 升级到「命名匹配 + 语义匹配」双层(规则表 + usage 推断),让 proposed_increment 真正具备判断力。 触发关键词:抽设计系统、提取设计 token、design token 提取、design.md 生成、扒站点设计、反向工程设计、tailwind config 生成、shadcn theme 生成、Figma 抽 token、Figma variables 导出、design system extract、design language、固化设计语言、Handoff design.md、给 AI 编程的设计上下文。 排除(反向):设计稿走查(用 /设计走查)、实现 vs 设计源差异核查(用 /设计验收)、竞品功能 / 交互模式拆解(用 /竞品拆解,那是模式层不是 token 层)、新建设计系统组件库本身(这是 SparkDesign 的工作,本 Skill 只产 design.md)、字体 / 图像 / SVG icon set 的版权下载(红线禁止)、Figma 组件几何形状提取(本 Skill 只抽 token,不抽 Auto Layout 结构)。
9 -
ahang1598 Bundle Linkfox Ruiguan Trademark Graphic Detection 2产品图片的图形商标检测与相似度搜索。当用户提到商标检测、图形商标搜索、Logo侵权检查、商标相似度分析、图片商标风险评估、产品图片商标筛查、graphic trademark detection, logo infringement, trademark similarity, trademark risk, image trademark screening, Ruiguan时触发此技能。即使用户未明确说"商标检测",只要其需求涉及将产品图片与已注册的图形商标进行比对或评估商标侵权风险,也应触发此技能。
9 -
johnalbertini14-glitch Bundle FontsChoose and implement web typography avoiding common rendering, pairing, and hierarchy mistakes.
1 -
johnalbertini14-glitch Bundle Happy HuesGet curated color palettes from Happy Hues for design projects. Use when the user asks for color schemes, palette recommendations, design colors, UI colors, or mentions "配色", "color palette", "happy hues", "颜色方案". Supports filtering by mood (warm, dark, bold, minimal, etc.) and provides role-based colors (background, headline, button, highlight, etc.).
1 -
johnalbertini14-glitch Bundle Canvas DesignCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
1 -
johnalbertini14-glitch Bundle Design System ComponentsPatterns for building design system components using Surface primitives, CVA variants, and consistent styling. Use when building reusable UI components that follow design token architecture. Triggers on Surface component, CVA, class-variance-authority, component variants, design tokens.
1 -
johnalbertini14-glitch Bundle Nano Banana Prompting SkillTransform natural language image requests into optimized structured prompts for Gemini image generation. Automatically detects style and builds the perfect prompt — cinematic, illustration, anime, 3D, watercolor, product, and more.
1
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 竞品拆解, 视觉情绪板, 数字人头像. 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.