Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计通过

tufte-slide-design簇绒幻灯片设计

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

480

周安装

20

GitHub Stars

7

下载量

160
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:tufte-slide-design(簇绒幻灯片设计)
来源仓库:https://github.com/ingpoc/skills
仓库路径:skills/tufte-slide-design
安装命令:
npx skills add https://github.com/ingpoc/skills --skill tufte-slide-design
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/ingpoc/skills --skill tufte-slide-design

简介

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。

  • 适合让 Agent 整理页面结构、生成 UI 方案或改进组件层级。
  • 需结合现有品牌、设计系统和用户任务使用,避免堆砌装饰元素。
  • 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和对齐。
  • tufte-slide-design 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Tufte Slide Design

Apply Edward Tufte's principles from "The Visual Display of Quantitative Information" to create presentations that communicate complex ideas with clarity, precision, and efficiency.

Core Philosophy

Tufte's central insight: "Clutter and confusion are failures of design, not attributes of information."

Information overload is rarely the problem—poor information design is. The goal is graphical excellence: the well-designed presentation of interesting data combining substance, statistics, and design.

The Five Laws of Data-Ink

When designing any slide with data:

  1. Above all else, show the data - Data is the primary focus
  2. Maximize the data-ink ratio - Every pixel should convey information
  3. Erase non-data-ink - Remove decorations that don't inform
  4. Erase redundant data-ink - Eliminate duplicate information carriers
  5. Revise and edit - Continuously refine toward simplicity

Data-Ink Ratio Formula

Data-Ink Ratio = Ink presenting data / Total ink used

Target: As close to 1.0 as possible. Each element should earn its place.

Slide Design Workflow

Step 1: Identify the Data Story

Before creating any slide, answer:

  • What is the ONE key insight this slide must communicate?
  • What data supports this insight?
  • What would be lost if this slide were removed?

Step 2: Apply the Chartjunk Elimination Checklist

Remove or minimize:

Chartjunk ElementAction
3D effectsFlatten to 2D
Gradient fillsUse solid colors
Heavy gridlinesLighten or remove
Decorative bordersRemove entirely
Background imagesRemove unless data
Drop shadowsRemove
Unnecessary legendsLabel directly on chart
Excessive tick marksReduce to minimum
Moiré patternsUse solid fills

Step 3: Check Graphical Integrity

Tufte's Six Principles of Graphical Integrity:

  1. Proportional representation - Visual size must match numerical quantity
  2. Clear labeling - Label data directly on the graphic
  3. Show data variation, not design variation - Design should not distort
  4. Use proper monetary units - Deflate/standardize when showing money over time
  5. Match dimensions - Don't use 2D/3D to represent 1D data
  6. Preserve context - Never quote data out of context

Step 4: Calculate the Lie Factor

Lie Factor = Size of effect in graphic / Size of effect in data
  • Lie Factor = 1.0: Truthful
  • Lie Factor > 1.0: Overstates the effect
  • Lie Factor < 1.0: Understates the effect

Example violation: A 53% numerical change shown as 783% visual change = Lie Factor of 14.8

Step 5: Apply Advanced Techniques

Small Multiples

Use for comparing related data:

  • Same graphic structure repeated with different data slices
  • Enables visual comparison within eye span
  • "Move to the heart of visual reasoning—to see, distinguish, choose"

Sparklines

Word-sized graphics for inline data display:

  • High resolution in small space
  • Embed in tables or text
  • "Datawords: data-intense, design-simple, word-sized graphics"

Direct Labeling

Instead of legends, label data directly:

  • Reduces eye movement
  • Eliminates legend decoding
  • Places information where attention focuses

Slide Types and Tufte Approaches

Data-Heavy Slides

  1. Strip unnecessary gridlines
  2. Use range-frame axes (only show data range)
  3. Consider small multiples for comparisons
  4. Direct label instead of legends
  5. Horizontal orientation where possible

Text-Heavy Slides (Anti-Pattern)

Tufte's critique of bullet points ("The Cognitive Style of PowerPoint"):

  • Bullet lists fragment thought
  • Hierarchical bullets obscure relationships
  • Low information density

Alternative approaches:

  • Use sentence-case prose for complex ideas
  • Provide detailed handouts instead
  • Show data tables with full context
  • Use visual diagrams showing relationships

Title Slides

Apply same principles:

  • Remove decorative elements
  • Use typography for hierarchy, not ornament
  • Every word should contribute meaning

Quick Reference: Before/After Patterns

Bar Charts

Before: 3D bars, gradient fills, heavy gridlines, legend below After: 2D bars, solid colors, no gridlines, direct labels

Line Charts

Before: Multiple colors, thick lines, point markers, legend After: Direct labels on lines, minimal markers, reduced palette

Pie Charts

Tufte's view: Generally avoid. If required:

  • Never use 3D
  • Limit to 3-4 slices maximum
  • Consider bar chart instead

Tables

Before: Heavy borders, alternating row colors, centered text After: Minimal rules, left-aligned text, whitespace for separation

Resources

For detailed principles and examples, reference:

  • references/tufte-principles.md - Complete principle documentation with examples
  • references/slide-checklist.md - Quick checklist for slide review

Anti-Patterns to Avoid

  1. PowerPoint defaults - Override all default templates and effects
  2. Chart templates - Design from data, not from template
  3. Decoration for engagement - Data is engaging when well-presented
  4. Hiding complexity - Show the complexity, design it well
  5. Animation for emphasis - Use visual hierarchy instead

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

39.06%
按下载量换算62

Claude

29.84%
按下载量换算48

Cursor

18.32%
按下载量换算29

Gemini CLI

8.9%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。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