tapcanvas-design-to-web

tapcanvas-design-to-web is a skill for Claude Code, Codex from anymouschina/TapCanvas. It costs 80 tokens per session (1,744 once invoked), scanned A, original, MIT.

A workflow for rebuilding a webpage from a provided design image into working frontend code. It first examines the image with a connected local vision system, then breaks down the layout and implements and compares the result.

In plain words
What is it for?
Use it to turn screenshots or design mockups into pages in an existing frontend project, including React and Mantine interfaces, with visual checks and corrections.
Why use it?
It helps developers reproduce a visual design accurately while keeping the result as a real, runnable page. It also prevents guessing when the required image-analysis workflow or input details are missing.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to turn screenshots or design mockups into pages in an existing frontend project, including React and Mantine interfaces, with visual checks and corrections.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/anymouschina/tapcanvas/tapcanvas-design-to-web
Install

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.

Any agent
npx skills add anymouschina/TapCanvas --skill tapcanvas-design-to-web
Clone the repo
git clone --depth 1 https://github.com/anymouschina/TapCanvas

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for tapcanvas-design-to-web

README.md
[![agentmods](https://agentmods.dev/badge/skills/anymouschina/tapcanvas/tapcanvas-design-to-web/github.svg)](https://agentmods.dev/skills/anymouschina/tapcanvas/tapcanvas-design-to-web)
Your own site
<a href="https://agentmods.dev/skills/anymouschina/tapcanvas/tapcanvas-design-to-web"><img src="https://agentmods.dev/badge/skills/anymouschina/tapcanvas/tapcanvas-design-to-web/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.

agentmods 80×15 button for tapcanvas-design-to-web

Your own site · 80×15
<a href="https://agentmods.dev/skills/anymouschina/tapcanvas/tapcanvas-design-to-web"><img src="https://agentmods.dev/badge/skills/anymouschina/tapcanvas/tapcanvas-design-to-web.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 80 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,744 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00080 $0.01744
Opus 5 $0.00040 $0.00872
Sonnet 5 $0.00016 $0.00349
Haiku 4.5 $0.00008 $0.00174

Measured 12d ago against content hash 60ea2935c7c0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

tapcanvas-design-to-web 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 12d 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.

apps/agents-cli/skills/tapcanvas-design-to-web/SKILL.md · 144 lines

How it starts

The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TapCanvas Design To Web

目标

  • 把用户提供的设计稿图片还原成仓库内真实可运行的网页,而不是只产出静态描述。
  • 优先追求接近设计稿的视觉完成度,同时保持实现符合当前仓库技术栈与模块边界。
  • 单一路径执行:设计稿取证 -> 结构拆解 -> 代码实现 -> 对稿修正。

强约束

  • 本 skill 强依赖本地已启动的 TapCanvas vision 工作流。
  • 默认视觉模型是 gemini-3.1-flash-image-preview;除非用户显式指定其他 active model,否则不要私自改。
  • 假定当前环境具备白名单调用权限;若无权限,必须直接报错。
  • 必须先做视觉取证,再动代码;禁止凭主观印象直接开写。
  • 本地 vision 工作流不可用、超时、返回空结果、资源不可达或证据不足时,必须显式报错并说明卡在哪一步。
  • 禁止静默降级到其他模型、默认视觉方案、硬编码模板或“差不多”的通用 landing page。
  • 只能使用用户本轮明确提供的设计稿路径、当前真实代码、当前项目状态和工具返回结果作为依据。

何时使用

  • 用户说“根据这张设计稿还原页面”“把截图做成网页”“参考这张 UI 图实现页面”。
  • 用户明确把设计稿图片放进仓库,例如 assets/,并要求还原为 React/Mantine 页面。
  • 用户要求“接近 AI Studio 做出的网页质感”,本质上是在追求高保真视觉落地。

必要输入

  • 至少一张设计稿图片路径。
  • 明确目标页面或落地区域:新页面、现有路由、现有组件改造,或其他清晰的前端 surface。
  • 明确还原标准:
    • 高保真还原
    • 保留风格但允许工程化调整
  • 确认本次任务应通过本地 TapCanvas vision 工作流做视觉分析。
  • 可选外部 vision prompt;若调用方提供,必须优先使用该 prompt,而不是静默改写。

缺少上述信息时,先指出缺失项,不要假设页面归属或交互细节。

视觉取证步骤

1. 收集输入事实

  • 读取用户给出的设计稿路径。
  • 确认要修改的页面入口、路由、组件或容器。
  • 如果是现有页面改造,先读现有实现,确认哪些结构可以保留,哪些必须重写。

2. 调用图像理解

优先使用本地 TapCanvas vision 工作流,对设计稿做结构化取证。目标不是“看图讲故事”,而是提取可编码的界面事实。

建议分析维度:

  • 页面结构:section 划分、主次层级、左右布局、卡片/列表/表单/导航的组合关系
  • 视觉 token:主色、辅助色、背景层次、圆角、边框、阴影、模糊、间距节奏
  • Typography:字号层级、字重、大标题风格、正文密度、按钮文字风格
  • 组件清单:hero、navbar、feature cards、metric tiles、gallery、CTA、form、footer
  • 装饰系统:渐变、光斑、噪点、网格、插画、mock device、浮层、标签
  • 交互暗示:hover、选中态、滚动分段、粘性头部、入场动画线索

调用时默认要求:

  • 默认 modelAlias="gemini-3.1-flash-image-preview"
  • 温度保持低值,例如 0.2
  • 若调用方提供外部 vision prompt,则直接透传该 prompt
  • 若未提供外部 vision prompt,才使用本 skill 内置的结构化分析提示
  • 若外部 prompt 与要求的输出结构冲突,必须显式报冲突,不能自己猜着改
  • 输出必须是结构化结果,至少包含:
    • layout
    • designTokens
    • components
    • visualHierarchy
    • motionHints
    • uncertainties

如果设计稿有多个画板或多张图片:

  • 先分别取证,再合并出统一 token 与组件契约。
  • 对互相冲突的部分显式标注,不要自行拍板“平均化”。

实现步骤

3. 把取证结果转成代码计划

先形成一个简短实施计划,再开始改代码:

  • 页面拆成哪些组件
  • 哪些 token 放在页面级样式,哪些放在复用样式
  • 哪些视觉效果要用纯 CSS,哪些需要少量交互动效
  • 哪些地方必须做响应式重排

坚持单一职责:

  • 页面容器负责整体结构
  • 各 section 组件只负责自己的视图
  • 视觉 token 与纯转换逻辑尽量抽成独立模块

4. 在仓库中落地

  • 优先改真实页面,不要只生成示例文件。
  • 遵守当前仓库技术栈:React + TypeScript + Mantine。
  • 每个 JSX/TSX 标签都要带可读 className
  • 不新增 any,不写 as any
  • 如果现有文件已经过大,先拆分再实现。
  • 如需补充图片资源,只能使用用户明确提供的设计稿或仓库已有合法资源;不能伪造未提供素材。

Read the full file on GitHub · 144 lines

Changes

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.

  1. 12d ago First seen · 144 lines · 80 tokens per session scan A 60ea2935c7c0

Subscribe to this mod's changes

tapcanvas-design-to-web is a skill published in the GitHub repository anymouschina/TapCanvas (603 stars, last pushed yesterday), licensed MIT. It adds 80 tokens to every session and 1,744 once invoked, about $0.0004 per session on Opus 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.

Related

Other skills, from other repositories

canvas-design

Design philosophy docs and canvas-based visual creation. Use when articulating design principles, crafting multi-page design documents, or exploring aesthetic philosophy with intentional design thinking.

PracticalSwan/agent-skills · 34 tokens

agentic-webdesign-framesmith

Use Framesmith MCP server to give AI agents a visual design canvas — sketch UI, review in browser, agree on design before writing framework code. Scene graph → HTML/CSS → PNG rendering with evaluation and auto-fix.

S3YED/appie-kit · 53 tokens

ui-styling

Create beautiful, accessible user interfaces with shadcn/ui components (built on Radix UI + Tailwind), Tailwind CSS utility-first styling, and canvas-based visual designs. Use when building user interfaces, implementing design systems, creating responsive layouts, adding accessible components (dialogs, dropdowns…

oyi77/1ai-skills · 90 tokens

cursor-reveal-hero

Build, repair, tune, and validate interactive cursor-trail mask reveal hero effects in frontend projects. Use when implementing canvas-based base/reveal compositing, soft radial cursor masks, fading cursor trails, reveal debug controls, full-screen hero scenes without site text, or browser smoke checks for cursor…

sergekostenchuk/mouse-trail-masking-reveal · 80 tokens

chakra-ui-builder

Build responsive, accessible UI components and layouts using Chakra UI v3, install or configure Chakra UI in new and existing projects, and design scalable themes using tokens, semantic tokens, recipes, and slot recipes. Use this skill whenever a user asks to build, create, or generate any UI component, page, form…

chakra-ui/chakra-ui · 214 tokens

frontend-visual-qa

Audits already-rendered web, landing-page, HTML deck/slide, browser tool/game, dashboard/admin, design-system, and desktop UIs using real-browser or native-app journeys, inspected screenshots, DOM geometry, responsive or projection viewports, and a bundled Playwright sweep. Use after UI implementation to find…

daymade/claude-code-skills · 145 tokens