Guides an AI agent through a two-phase web design workflow: create and confirm a DESIGN.md specification, then generate responsive UI code for landing pages, portfolios, product pages, blogs, and SaaS sites.
Web 视觉设计 SKILL。输入 PRD / 参考 URL / 截图 / 关键词（任意组合），先产出一份标准化 DESIGN.md 设计规范，用户确认后据此生成 UI/UX、视觉、动效、响应式全部达标的 web 代码。专攻 web 端：Landing Page、Portfolio、产品页、博客、个人站、SaaS 介绍页等。当用户说"帮我做个网站""设计一个页面""参考 XX 做一个""把这个截图/PRD 做成网页""做一个 landing page""出一份 design 规范"时触发。不用于后端、数据库、纯逻辑 bug 修复。
Generates illustrations for Chinese deep-dive articles using fixed IP characters, visual metaphors, shot lists, selected art styles, and image-generation APIs.
Creates a reusable minimalist hand-drawn IP character from authorized user materials, then plans and produces illustrated content, knowledge cards, posters, storyboards, and prese…
Generates and edits Xiaohongshu (XHS/RedNote) cover images from photos, titles, and style instructions, with GPT Image 2/Codex as the primary method and Gemini CLI as fallback.
Creates polished, shareable HTML pages from ideas, files, folders, URLs, datasets, and exports, using automatic routing to specialized design systems and interactive layouts.
Defines user-facing product behavior from validated evidence and approved scope, including information architecture, task flows, state models, interaction contracts, usability pla…
Provides guidance and scripts for ICC profile inspection, color-space conversion, gamut analysis, working-space validation, soft proofing, and color workflow troubleshooting.
Guides agents in documenting brand strategy, visual identity, voice, applications, governance, and assets using templates, validation, compilation, and optional image prompt gener…
Guides agents in creating subject-neutral images and videos with a near-monochrome analog occultism and industrial CRT noir aesthetic, including lighting, composition, signal degr…
