Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add chuancyzhang/deepseek-cowork --skill huashu-designgit clone --depth 1 https://github.com/chuancyzhang/deepseek-coworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/chuancyzhang/deepseek-cowork/huashu-design)<a href="https://agentmods.dev/skills/chuancyzhang/deepseek-cowork/huashu-design"><img src="https://agentmods.dev/badge/skills/chuancyzhang/deepseek-cowork/huashu-design/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/chuancyzhang/deepseek-cowork/huashu-design"><img src="https://agentmods.dev/badge/skills/chuancyzhang/deepseek-cowork/huashu-design.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00100 | $0.15757 |
| Opus 5.5 | $0.00040 | $0.06303 |
| Sonnet 5.5 | $0.00020 | $0.03151 |
| Haiku 4.5 | $0.00010 | $0.01576 |
Grade A, and why
huashu-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- visual-deck-studio — 88% identical, 29 lines differ
How it starts
The opening of the file, as written. The whole thing — 501 lines — stays where its author put it; the contents beside it link to each section on GitHub.
花叔Design · Huashu-Design
你是一位用HTML工作的设计师，不是程序员。用户是你的manager，你产出深思熟虑、做工精良的设计作品。
HTML是工具，但你的媒介和产出形式会变——做幻灯片时别像网页，做动画时别像Dashboard，做App原型时别像说明书。根据任务embody对应领域的专家：动画师/UX设计师/幻灯片设计师/原型师。
使用前提
这个skill专为「用HTML做视觉产出」的场景设计，不是给任何HTML任务用的万能勺。适用场景：
- 交互原型：高保真产品mockup，用户可以点击、切换、感受流程
- 设计变体探索：并排对比多个设计方向，或用Tweaks实时调参
- 演示幻灯片：1920×1080的HTML deck，可以当PPT用
- 动画Demo：时间轴驱动的motion design，做视频素材或概念演示
- 信息图/可视化：精确排版、数据驱动、印刷级质量
不适用场景：生产级Web App、SEO网站、需要后端的动态系统——这些用frontend-design skill。
任务路由：一张表定入口
收到任务先扫一遍这张表，确定走哪条线再开工（多信号同时命中按行序叠加）：
| 任务信号 | 入口 |
|---|---|
| 提到具体品牌/产品名 | 核心原则#0 事实验证 → §1.a 资产协议 → 标准流程 |
| 没给风格参考（最常见） | Fallback 顾问模式 Phase 1-5 → 回标准流程 Step 2 |
| 幻灯片/PPT | 标准流程 + Step 1 deck 交付链 + 「技术红线」架构选型 |
| 动画/导出 MP4/GIF | 标准流程 + Step 9；动手前必读 references/animation-pitfalls.md |
| 带解说长视频（≥1分钟） | Step 9.5 → references/voiceover-pipeline.md |
| launch film/品牌宣传片（「Apple级」「超级碗品质」） | 先写万字 director's notes → references/launch-film-director-notes.md |
| App/iOS 原型 | 「App / iOS 原型专属守则」（覆盖通用规则） |
| 评审/打分 | Step 10 → references/critique-guide.md |
| 弱 runtime（无 subagent/非 Claude） | 上述任一条 + 「弱 runtime 降级模式」 |
例：「做个咖啡主题的 PPT」= 第 2 行 + 第 3 行——Fallback 出三版（咖啡是主题不是品牌，不找 logo），deck 骨架统一用概览墙模板。
核心原则 #0 · 事实验证先于假设（优先级最高，凌驾所有其他流程）
任何涉及具体产品/技术/事件/人物的存在性、发布状态、版本号、规格参数的事实性断言，第一步必须
WebSearch验证，禁止凭训练语料做断言。
触发条件（满足任一）：
- 用户提到你不熟悉或不确定的具体产品名（如"大疆 Pocket 4"、"Nano Banana Pro"、"Gemini 3 Pro"、某新版 SDK）
- 涉及 2024 年及之后的发布时间线、版本号、规格参数
- 你内心冒出"我记得好像是..."、"应该还没发布"、"大概在..."、"可能不存在"的句式
- 用户请求给某个具体产品/公司做设计物料
硬流程（开工前执行，优先于 clarifying questions）：
WebSearch产品名 + 最新时间词（"2026 latest"、"launch date"、"release"、"specs"）- 读 1-3 条权威结果，确认：存在性 / 发布状态 / 最新版本号 / 关键规格
- 把事实写进项目的
product-facts.md（见工作流 Step 2），不靠记忆 - 搜不到或结果模糊 → 问用户，而不是自行假设
反例（2026-04-20 实测）：用户要「大疆 Pocket 4 发布动画」，我凭记忆断言「还没发布」做了概念剪影——真相是 4 天前已发布、官方物料俱在。成本对比：WebSearch 10 秒 << 返工 2 小时。
这条原则优先级高于"问 clarifying questions"——问问题的前提是你对事实已有正确理解。事实错了，问什么都是歪的。
禁止句式（看到自己要说这些时，立即停下去搜）：
- ❌ "我记得 X 还没发布"
- ❌ "X 目前是 vN 版本"（未经搜索的断言）
- ❌ "X 这个产品可能不存在"
- ❌ "据我所知 X 的规格是..."
- ✅ "我
WebSearch一下 X 最新状态" - ✅ "搜到的权威来源说 X 是 ..."
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/android_frame.jsx 4.4 KB
- assets/animations.jsx 10 KB
- assets/banner.svg 9.3 KB
- assets/browser_window.jsx 3.9 KB
- assets/deck_index.html 21 KB
- assets/deck_stage.js 11 KB runs code
- assets/design_canvas.jsx 5.1 KB
- assets/ios_frame.jsx 4.6 KB
- assets/macos_window.jsx 2.5 KB
- assets/narration_stage.jsx 18 KB
- assets/personal-asset-index.example.json 1.8 KB
- demos/c1-ios-prototype-en.html 34 KB
- demos/c1-ios-prototype.html 34 KB
- demos/c2-slides-pptx-en.html 32 KB
- demos/c2-slides-pptx.html 32 KB
- demos/c3-motion-design-en.html 36 KB
- demos/c3-motion-design.html 36 KB
- demos/c4-tweaks-en.html 30 KB
- demos/c4-tweaks.html 30 KB
- demos/c5-infographic-en.html 24 KB
- demos/c5-infographic.html 24 KB
- demos/c6-expert-review-en.html 26 KB
- demos/c6-expert-review.html 26 KB
- demos/hero-animation-v10-en.html 48 KB
- demos/md-html-narration/md-html-demo.html 36 KB
- demos/md-html-narration/script.md 3.4 KB
- demos/voiceover-demo/script.md 589 B
- demos/voiceover-demo/什么是token.html 11 KB
- demos/w1-brand-protocol-en.html 20 KB
- demos/w1-brand-protocol.html 20 KB
- demos/w2-junior-designer-en.html 31 KB
- demos/w2-junior-designer.html 31 KB
- demos/w3-fallback-advisor-en.html 23 KB
- demos/w3-fallback-advisor.html 26 KB
- LICENSE 1.1 KB
- package-lock.json 28 KB
- package.json 133 B
- README.en.md 17 KB
- README.md 18 KB
- references/animation-best-practices.md 21 KB
- references/animation-pitfalls.md 25 KB
- references/animations.md 7.3 KB
- references/app-prototype.md 9.7 KB
- references/apple-gallery-showcase.md 11 KB
- references/audio-design-rules.md 9.4 KB
- references/brand-asset-protocol.md 15 KB
- references/cinematic-patterns.md 11 KB
- references/content-guidelines.md 8.2 KB
- references/critique-guide.md 8.8 KB
- references/design-context.md 6.5 KB
- references/design-styles.md 45 KB
- references/editable-pptx.md 16 KB
- references/hero-animation-case-study.md 11 KB
- references/launch-film-director-notes.md 14 KB
- references/multi-perspective-parallel-case-study.md 11 KB
- references/react-setup.md 9.1 KB
- references/scene-templates.md 7.0 KB
- references/sfx-library.md 9.5 KB
- references/slide-decks.md 35 KB
- references/tweaks-system.md 8.8 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 501 lines · 100 tokens per session scan A 614420f75170
huashu-design is a skill published in the GitHub repository chuancyzhang/deepseek-cowork (18 stars, last pushed today), licensed Apache-2.0. It adds 100 tokens to every session and 15,757 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ultragoal with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
glance-canonical-layouts
Use this skill to create home screen widgets using Jetpack Glance Canonical Layouts from the Android Platform Samples project, or when asked to implement standard widget layout patterns (including Check List, Image Grid, Snap Scrolling, and Standard Toolbar).
accessibility
Consolidated accessibility skill entrypoint for WCAG 2.2, ARIA Authoring Practices, cognitive accessibility, Section 508, EN 301 549, design intent verification, and the Accessibility Planner workflow.