Token导航 LogoToken导航TokenDH.com
前端设计操作浏览器github未标认证来源可访问许可证需确认审计提醒

ui-wireframe-generatorui 线框生成器

Agent Skill

用于辅助界面设计、视觉规范、排版、配色、布局和交互体验优化。它适合让 Agent 根据产品场景整理页面结构、生成 UI 方案、检查视觉一致性或改进组件层级。使用时需要结合现有品牌、设计系统和用户任务,不应只堆装饰元素;涉及真实页面改动时,应通过截图或浏览器预览检查文本溢出、对齐和响应式表现。

总安装

451

周安装

19

GitHub Stars

6

下载量

158
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:ui-wireframe-generator(ui 线框生成器)
来源仓库:https://github.com/enlinks-llc/katsuragi
仓库路径:skills/ui-wireframe-generator
安装命令:
npx skills add https://github.com/enlinks-llc/katsuragi --skill ui-wireframe-generator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/enlinks-llc/katsuragi --skill ui-wireframe-generator

简介

用于辅助界面设计、视觉规范和交互体验优化。ui-wireframe-generator 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

  • 适合整理页面结构、生成 UI 方案或检查视觉一致性。
  • 使用时需结合品牌、设计系统和用户任务,避免堆砌装饰元素。
  • 涉及页面改动时应通过截图或浏览器预览检查文本溢出和对齐。
  • 建议配合现有设计系统使用,确保响应式表现符合预期。

SKILL.md

Katsuragi UI Wireframe Generator

Generate text-based UI wireframes using .kui file syntax. Katsuragi converts simple declarative files into SVG/PNG images, serving as a communication medium for UI discussions.

GitHub: https://github.com/enlinks-llc/katsuragi

Install

# Install this skill
npx skills add enlinks-llc/katsuragi

# Install katsuragi CLI
npm install -g katsuragi
# or use directly with npx
npx ktr input.kui -o output.png

Why Katsuragi?

  • LLM-friendly: Text-based format that AI agents can read, write, and iterate on
  • Human-readable: Simple syntax anyone can understand without design tools
  • Git-friendly: Version control your wireframes alongside code
  • Zero design skills needed: Grid-based layout handles positioning automatically
  • Instant output: Generate PNG/SVG in seconds via CLI

Use Cases

  • Developer ↔ Stakeholder communication: Show rough UI ideas without Figma
  • AI-assisted UI development: Let LLMs propose layouts in .kui format
  • Quick prototyping: Sketch screen layouts before writing any frontend code
  • Documentation: Embed wireframes in technical specs and PRs
  • Design discussions: Iterate on layouts through text diffs
  • Reverse engineering: Convert existing webpages to .kui wireframes with ktr fetch

When to Use

  • Creating quick UI wireframes or mockups
  • Discussing UI layouts with stakeholders
  • Generating visual representations of UI designs
  • Documenting UI structures in a text-based format

Quick Start

To generate a wireframe:

npx ktr input.kui -o output.png  # PNG output
npx ktr input.kui -o output.svg  # SVG output

Fetch Command

Convert existing webpages to .kui wireframes:

# Basic usage
npx ktr fetch https://example.com -o output.kui

# With viewport option (desktop, mobile, tablet)
npx ktr fetch https://example.com -o output.kui --viewport mobile

# With custom grid and ratio
npx ktr fetch https://example.com -o output.kui --grid 6x4 --ratio 4:3

Options

OptionDescriptionDefault
-o, --output <file>Output.kui file path (required)-
--viewport <type>Target viewport: desktop, mobile, tabletdesktop
--grid <grid>Override grid size (e.g., 4x3)auto
--ratio <ratio>Override aspect ratio (e.g., 16:9)auto

How It Works

  1. Fetches HTML from the URL (no headless browser)
  2. Extracts <meta name="theme-color"> as the primary color for the colors header
  3. Extracts inline style colors (background-color, border-color) per element
  4. Parses DOM to extract visual elements (header, nav, main, footer, buttons, inputs, etc.)
  5. Auto-calculates optimal grid size based on element positions
  6. Maps HTML elements to kui components with real content:

- header, nav, footer, etc. → txt with semantic labels ("Header", "Navigation", "Footer") - buttonbtn with actual button text - input, textarea, selectinput with placeholder or type-based label - imgimg with alt text - h1-h6, p, span, label (with text) → txt with actual text content (truncated)

  1. Applies extracted colors to component bg and border properties
  2. Generates .kui file with cell assignments

Limitations

  • No CSS parsing (structure-based analysis only)
  • No JavaScript execution (SPA not supported)
  • Maximum grid size: 26×26
  • Text content is truncated for wireframe display (max 20-30 characters)

File Structure

Every .kui file follows this structure:

ratio: 16:9              // Canvas aspect ratio
grid: 4x3                // Grid dimensions (columns x rows)
gap: 8                   // Space between cells (optional)
padding: 16              // Canvas padding (optional)
colors: { name: "#hex" } // Color definitions (optional)

// Cell definitions
A1..B2: { type: component, property: value }

Grid System

Cells use Excel-style notation:

  • Single cell: A1 (column A, row 1)
  • Range: A1..B2 (merges cells from A1 to B2)
  • Columns: A, B, C, D... (left to right)
  • Rows: 1, 2, 3... (top to bottom)

Grid divides canvas equally. A 4x3 grid creates 4 equal columns and 3 equal rows.

Components

TypePurposeKey Properties
txtText labelvalue, align
boxEmpty containerbg, border
btnButtonvalue, bg
inputForm inputlabel
imgImage placeholderalt

Common Properties

PropertyValuesDefault
alignleft, center, rightleft
bgColor (hex, name, or $ref)#e0e0e0
borderColor (hex, name, or $ref)none
paddingNumber (pixels)0

Color System

Define reusable colors in the header:

colors: { primary: "#3B82F6", danger: "#EF4444", accent: "orange" }

// Use with $ prefix
D3: { type: btn, value: "Submit", bg: $primary }

Supported formats: #RGB, #RRGGBB, CSS color names

Output Specifications

  • Longest edge: Fixed at 1280px
  • Aspect ratios: 16:9 → 1280×720, 4:3 → 1280×960, 9:16 → 720×1280

Syntax Reference

For complete syntax details, see references/syntax.md.

Examples

Login Form

ratio: 16:9
grid: 4x3
gap: 8
padding: 16
colors: { primary: "#3B82F6" }

// Header
A1..D1: { type: txt, value: "Login", align: center }

// Form fields
A2..D2: { type: input, label: "Email" }
A3..C3: { type: input, label: "Password" }

// Submit button
D3: { type: btn, value: "Login", bg: $primary }

Dashboard Layout

ratio: 16:9
grid: 4x3
gap: 8
padding: 16
colors: { secondary: "#f5f5f5", outline: "#333" }

// Navigation
A1: { type: img, alt: "Logo" }
B1..C1: { type: txt, value: "Dashboard", align: center }
D1: { type: btn, value: "Logout", border: $outline }

// Sidebar
A2..A3: { type: box, bg: $secondary }

// Main content area
B2..D2: { type: txt, value: "Welcome back!", align: left }
B3..D3: { type: box, border: $outline }

Mobile Profile (Vertical)

ratio: 9:16
grid: 3x5
gap: 8
padding: 16
colors: { primary: "#3B82F6", outline: "#333" }

// Header
A1..C1: { type: txt, value: "My App", align: center }

// Profile image
A2..C2: { type: img, alt: "Profile Photo" }

// User info
A3..C3: { type: txt, value: "John Doe", align: center }

// Actions
A4..C4: { type: btn, value: "Edit Profile", bg: $primary }
A5..C5: { type: btn, value: "Settings", border: $outline }

Workflow

  1. Define the canvas - Set ratio and grid size based on target layout
  2. Plan the grid - Sketch cell positions mentally or on paper
  3. Write.kui file - Start with header, then define cells top-to-bottom
  4. Generate output - Run npx ktr file.kui -o output.png
  5. Iterate - Adjust grid, spacing, or components as needed

Best Practices

  • Use meaningful color names in the colors block
  • Add comments (//) to organize sections
  • Start with larger merged cells for main areas, then refine
  • Use gap for clean separation between elements
  • Test with both SVG and PNG to choose the best format

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.36%
按下载量换算61

Claude

28.77%
按下载量换算45

Cursor

16.61%
按下载量换算26

Gemini CLI

10.26%
按下载量换算16

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills