AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Catch Web

skill-yushengxin-catch-web-catch-web · by yushengxin

>-

No reviews yet
0 installs
34 views
0.0% view→install

Install

$ agentstack add skill-yushengxin-catch-web-catch-web

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-yushengxin-catch-web-catch-web)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Catch Web? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

catch-web

从「好看的参考设计」到「自己可用的同风格页面」的两阶段工作流。

何时触发

  • 用户说"照着这个截图/网页做一版"
  • 用户说"提取这个设计风格"
  • 用户说"我要这个网站的风格"
  • 用户给一张图或 URL,并要求基于此生成 HTML

工作流

整个 skill 只有两步,没有"先讲 schema"这种没必要的步骤。

Phase 1: Extract — 提取 DNA

输入:一张或多张参考图 / 一个或多个 URL 输出:一份 dna.json 文件,按 [references/schema.md](references/schema.md) 的两层结构

步骤

  1. 读取 [references/analysis-guide.md](references/analysis-guide.md) 了解提取方法
  2. 读取 [references/schema.md](references/schema.md) 了解输出结构
  3. 核心层(必填) 8 个区块全部填实:color / typography / spacing / layout / shape / elevation / motion / mood
  4. 效果层(按需) 只在参考里实际观察到效果时才填
  5. 把结果写到 dna.json,每个字段都不能留空字符串,没观察到的用 null + 在 notes 里说明

多参考冲突处理:以"主导模式"为准,notes 数组里记录次要变体。

Phase 2: Generate — 生成 HTML

输入dna.json + 用户自己的内容(标题、段落、列表、图片等) 输出:自包含单文件 HTML(inline CSS + JS,无外部构建)

步骤

  1. 读取 [references/generation-guide.md](references/generation-guide.md) 了解代码生成规则
  2. 把 DNA 核心层映射成 CSS 自定义属性写到 :root
  3. 按效果层选择合适的技术(CSS / Canvas / WebGL / GSAP)
  4. 用户内容按 DNA 风格重新排版
  5. 单文件交付,所有 CSS/JS inline,CDN 仅用于大型库(three.js/gsap/lottie)

输入格式

  • 图片:用户直接附图,或提供本地路径
  • URL:用户给完整 URL,AI 调用 web_fetch 抓取页面分析
  • 多个参考:可以,AI 会综合判断"主导风格"

输出格式

  • DNA:JSON 文件,UTF-8,缩进 2 空格
  • HTML:单文件 .html,inline ` + `,可直接双击在浏览器打开

重要规则

  1. 不要照抄参考内容——只借风格,不借文字/图片
  2. 效果层用 null 代替猜测——没看见就别编
  3. CDN 选 jsdelivr——https://cdn.jsdelivr.net/npm/@/...,固定版本号,不写 @latest
  4. 自包含优先——能 inline 就 inline,少一次网络请求
  5. 可访问性——保证 WCAG AA 对比度(正文 4.5:1,大字 3:1)

详细规范

  • 字段定义:[references/schema.md](references/schema.md)
  • 提取方法:[references/analysis-guide.md](references/analysis-guide.md)
  • 代码生成:[references/generation-guide.md](references/generation-guide.md)
  • 示例:[examples/](examples/)

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.