Token导航 LogoToken导航

Agent Skill · 研究检索

design-thinking(设计思维)

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。

design-thinking 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 23,167。

总安装 4,348周安装 183GitHub Stars 61下载 1,523核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

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:design-thinking(设计思维)
来源仓库:https://github.com/melodic-software/claude-code-plugins
仓库路径:skills/design-thinking
安装命令:
npx skills add https://github.com/melodic-software/claude-code-plugins --skill design-thinking
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/melodic-software/claude-code-plugins --skill design-thinking

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

采用斯坦福 d.school 五步法推动以人为中心的产品创新。适用于需求分析、问题定义、创意发散到原型测试的全周期。提供工作坊模板与练习工具,支持远程协作与线下引导。需明确目标用户与业务约束,避免空泛讨论或无边界探索。design-thinking 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/design-thinking

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/melodic-software/claude-code-plugins --skill d…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

design-thinking 怎么安装?

在终端执行 npx skills add https://github.com/melodic-software/claude-code-plugins --skill design-thin…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

design-thinking 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

design-thinking 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

design-thinking 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

design-thinking 用的人多吗?

公开数据是累计安装 4,348 次,近一周安装 183 次,下载 1,523 次,GitHub 61 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

design-thinking 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
design-thinking

设计思维 · 本页

4,34861需要联网MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Design Thinking

When to Use This Skill

Use this skill when:

  • Design Thinking tasks - Working on design thinking methodology for human-centered innovation. covers the 5-phase ideo/stanford d.school approach (empathize, define, ideate, prototype, test) with workshop facilitation and exercise templates
  • Planning or design - Need guidance on Design Thinking approaches
  • Best practices - Want to follow established patterns and standards

Overview

A human-centered approach to innovation that integrates the needs of people, the possibilities of technology, and the requirements for business success.

What is Design Thinking?

Design Thinking is a problem-solving methodology developed by IDEO and taught at Stanford d.school. It emphasizes understanding users, challenging assumptions, redefining problems, and creating innovative solutions through iterative prototyping and testing.

Core Principles

PrincipleDescription
Human-CenteredStart with people, not technology or process
Embrace AmbiguityComfortable with uncertainty early on
Bias to ActionLearn by doing, not just thinking
Radical CollaborationCross-functional, diverse perspectives
Iterate RapidlyFail fast, learn fast, improve fast

The 5 Phases

┌─────────┐  ┌─────────┐  ┌─────────┐  ┌─────────┐  ┌─────────┐
│         │  │         │  │         │  │         │  │         │
│EMPATHIZE│→│ DEFINE  │→│ IDEATE  │→│PROTOTYPE│→│  TEST   │
│         │  │         │  │         │  │         │  │         │
│Understand│  │Frame the│  │Generate │  │Build to │  │Learn &  │
│the user │  │problem  │  │ideas    │  │learn    │  │refine   │
└─────────┘  └─────────┘  └─────────┘  └─────────┘  └─────────┘
     ▲                                                    │
     └────────────────── Iterate ─────────────────────────┘

Phase Characteristics

PhaseModeKey QuestionOutput
EmpathizeDivergeWhat is the user experiencing?Insights, observations
DefineConvergeWhat is the core problem?POV statement
IdeateDivergeWhat are possible solutions?Ideas, concepts
PrototypeConvergeHow can we make it tangible?Prototypes
TestLearnWhat works and what doesn't?Validated learning

Phase 1: Empathize

Build deep understanding of users and their experiences.

Empathy Methods

MethodDurationParticipantsBest For
User Interviews30-60 min1:1Deep individual insight
ObservationHours-daysField workNatural behavior
Empathy Mapping15-30 minWorkshopSynthesizing research
Journey Mapping1-2 hoursWorkshopExperience visualization
ShadowingHalf-dayField workWorkflow understanding

Empathy Map Template

## Empathy Map: [User/Persona Name]

### Says
[Direct quotes from interviews/observation]
- "[Quote 1]"
- "[Quote 2]"

### Thinks
[What might they be thinking? Hopes/fears?]
- [Thought 1]
- [Thought 2]

### Does
[Actions and behaviors observed]
- [Behavior 1]
- [Behavior 2]

### Feels
[Emotional state, what worries/excites them]
- [Emotion 1]
- [Emotion 2]

### Pains
[Frustrations, obstacles, risks]
- [Pain 1]
- [Pain 2]

### Gains
[Wants, needs, measures of success]
- [Gain 1]
- [Gain 2]

Interview Guide Structure

## User Interview Guide

**Duration:** 45-60 minutes
**Objective:** [What we want to learn]

### Opening (5 min)
- Introduce yourself and purpose
- Explain confidentiality, recording consent
- "There are no right or wrong answers"

### Context (10 min)
- Tell me about your role/day-to-day
- How long have you been doing [X]?
- Walk me through a typical [scenario]

### Deep Dive (25 min)
- Tell me about a time when [specific experience]
- What was the most challenging part?
- What did you try? What happened?
- How did that make you feel?

### Needs & Desires (10 min)
- If you could wave a magic wand, what would change?
- What would that enable you to do?
- What does success look like for you?

### Closing (5 min)
- Is there anything else you'd like to share?
- Can we follow up if we have questions?

Phase 2: Define

Synthesize observations and frame the right problem.

Point of View (POV) Statement

## POV Statement

**User:** [Specific user type/persona]

**Need:** [Verb - what they need to do]

**Insight:** [Because - surprising insight about why]

---

**Full Statement:**
[User] needs to [need] because [insight].

**Example:**
A busy working parent needs to quickly prepare healthy meals
because they want to model good eating habits for their children
but feel guilty about the time-cost tradeoff.

How Might We (HMW) Questions

Transform POV into actionable opportunity statements:

## How Might We Questions

**POV:** [Your POV statement]

### HMW Brainstorm

Generate HMW questions by asking:

| Lens | Question | HMW |
|------|----------|-----|
| **Amplify the good** | What's working? | HMW do more of [good thing]? |
| **Remove the bad** | What's painful? | HMW eliminate [pain point]? |
| **Explore the opposite** | What if we flipped it? | HMW [opposite approach]? |
| **Question assumptions** | What's assumed? | HMW if [assumption] wasn't true? |
| **Change status quo** | What's expected? | HMW change when/where/who? |
| **Break it into parts** | What are components? | HMW address just [component]? |

### Selected HMW Questions
1. HMW [question 1]?
2. HMW [question 2]?
3. HMW [question 3]?

**Priority HMW:** [The one to ideate on]

Phase 3: Ideate

Generate a wide range of creative solutions.

Ideation Rules

RuleDescription
Defer judgmentNo criticism during brainstorming
Go for quantityMore ideas = better ideas
Build on others' ideas"Yes, and..."
Encourage wild ideasThink big, no constraints
Be visualSketch, don't just talk
One conversationListen, build, riff
Stay focusedKeep the HMW visible

Brainstorming Session Template

## Ideation Session

**HMW:** [The How Might We question]
**Duration:** 30-45 minutes
**Participants:** [Names]

### Warm-Up (5 min)
[Quick creative exercise to loosen up]

### Silent Brainstorm (10 min)
- Write one idea per sticky note
- Work individually
- Aim for 10+ ideas each

### Share & Build (15 min)
- Share ideas one by one
- Build on others' ideas
- No criticism, only "Yes, and..."

### Cluster & Theme (10 min)
- Group similar ideas
- Name each cluster
- Identify promising directions

### Ideas Generated
| Cluster | Ideas | Count |
|---------|-------|-------|
| [Theme 1] | [List] | [N] |
| [Theme 2] | [List] | [N] |

### Top Ideas to Prototype
1. [Idea 1]
2. [Idea 2]
3. [Idea 3]

Ideation Techniques

TechniqueDescriptionWhen to Use
Classic BrainstormFree associationStarting point
BrainwritingSilent written ideasQuiet/remote groups
SCAMPERSubstitute, Combine, Adapt, Modify, Put to other use, Eliminate, ReverseImproving existing
AnalogiesHow would X solve this?Breaking mental models
Worst IdeaIntentionally bad ideasWhen stuck
Mash-upCombine unrelated conceptsRadical innovation

Phase 4: Prototype

Build quick, cheap representations to learn.

Prototype Philosophy

"If a picture is worth 1,000 words, a prototype is worth 1,000 meetings." — IDEO

Prototype Types

TypeFidelityTimeBest For
Paper SketchVery LowMinutesEarliest exploration
StoryboardLow30 minUser journey, scenarios
Paper PrototypeLow1-2 hoursScreen flows, UI
Role PlayLow30 minService experiences
Wizard of OzMediumHoursComplex interactions
Clickable MockupMediumDaysDigital products
Physical ModelVariesVariesTangible products

Prototype Planning Template

## Prototype Plan

**Concept:** [What we're prototyping]
**Goal:** [What we want to learn]
**Audience:** [Who will test it]

### What to Prototype
- **Include:** [Essential elements to test]
- **Exclude:** [What we won't build yet]

### Prototype Type
**Type:** [Paper/Digital/Physical/Role-play]
**Fidelity:** [Low/Medium/High]
**Time to Build:** [Duration]

### Materials Needed
- [Material 1]
- [Material 2]

### Key Questions to Answer
1. [Question 1]
2. [Question 2]
3. [Question 3]

### Success Criteria
- [How we'll know it worked]

Storyboard Template

## Storyboard: [Concept Name]

### Scene 1: [Setup]
[Sketch/Description]
**What's happening:** [Action]
**User says/thinks:** "[Quote]"

### Scene 2: [Trigger]
[Sketch/Description]
**What's happening:** [Action]
**User says/thinks:** "[Quote]"

### Scene 3: [Interaction]
[Sketch/Description]
**What's happening:** [Action]
**User says/thinks:** "[Quote]"

### Scene 4: [Resolution]
[Sketch/Description]
**What's happening:** [Action]
**User says/thinks:** "[Quote]"

Phase 5: Test

Learn what works, what doesn't, and iterate.

Testing Principles

PrincipleDescription
Show, don't tellLet the prototype speak
Create experiencesHave users interact, not just look
Ask "why?"Dig into reactions
Embrace failureEvery test teaches something
IterateBuild-test-learn cycles

Test Session Template

## Test Session

**Prototype:** [Name/Version]
**Participant:** [Code or name]
**Date:** [ISO date]

### Introduction
"We're testing a concept, not you. There are no wrong answers.
Please think out loud as you interact with this."

### Scenario
[Context to set up the test]

### Observations

| Moment | Observation | Quote | Insight |
|--------|-------------|-------|---------|
| [When] | [What happened] | "[Said]" | [Meaning] |

### Key Learnings
1. [Learning 1]
2. [Learning 2]
3. [Learning 3]

### What to Keep
- [Element that worked]

### What to Change
- [Element to modify]

### Questions Raised
- [New question to explore]

Iteration Framework

## Iteration Review

**Prototype Version:** [N]
**Tests Conducted:** [Number]

### Synthesis

| Finding | Frequency | Severity | Action |
|---------|-----------|----------|--------|
| [Finding] | [X/N] | H/M/L | Keep/Change/Remove |

### Next Iteration

**Priority Changes:**
1. [Change 1]
2. [Change 2]

**Hypothesis to Test:**
If we [change], then [expected outcome] because [reason].

Workshop Agenda Templates

Half-Day Workshop (4 hours)

## Design Thinking Workshop

**Duration:** 4 hours
**Participants:** 6-12 people

### Agenda

**0:00 - 0:15** Welcome & Icebreaker
- Introductions
- Creative warm-up

**0:15 - 0:45** Empathy Share (30 min)
- Share research/interviews
- Empathy mapping exercise

**0:45 - 1:15** Define (30 min)
- Synthesize insights
- Create POV statements
- Generate HMW questions

**1:15 - 1:30** Break (15 min)

**1:30 - 2:15** Ideate (45 min)
- Brainstorming
- Clustering
- Voting

**2:15 - 3:15** Prototype (60 min)
- Teams build low-fi prototypes
- Prepare to present

**3:15 - 3:45** Share & Feedback (30 min)
- Each team presents
- Group feedback

**3:45 - 4:00** Wrap-up (15 min)
- Key decisions
- Next steps

Full-Day Sprint (8 hours)

## Design Sprint (1-Day)

**Duration:** 8 hours (condensed from 5-day sprint)
**Participants:** 5-8 people cross-functional

### Morning: Understand & Define

**9:00 - 9:30** Set the stage
- Sprint goal
- Challenge statement

**9:30 - 10:30** Expert interviews / Lightning talks
- Stakeholder perspectives
- User insights

**10:30 - 11:00** How Might We
- Generate HMW from insights
- Vote on priority

**11:00 - 12:00** Sketch solutions
- Individual crazy 8s
- Solution sketches

### Afternoon: Build & Test

**12:00 - 1:00** Lunch

**1:00 - 1:30** Decide
- Critique sketches
- Vote
- Decide what to prototype

**1:30 - 4:00** Prototype
- Build testable prototype
- Prepare test script

**4:00 - 5:00** Test
- 3-5 user tests
- Capture feedback

**5:00 - 5:30** Debrief
- What we learned
- Next steps

Integration with Other Methods

Upstream

  • User Research - Feeds into Empathize phase
  • Stakeholder Analysis - Identifies who to interview
  • Strategic Analysis - Business context for innovation

Downstream

  • Lean Startup - Build-Measure-Learn loops
  • Agile Development - Iterative implementation
  • Business Model Canvas - Modeling the solution

.NET/C# Context

When applying Design Thinking to software products:

// Example: Tracking Design Thinking sessions
public class DesignThinkingSession
{
    public Guid Id { get; init; }
    public required string ChallengeName { get; init; }
    public required DesignPhase CurrentPhase { get; init; }
    public required string HowMightWe { get; init; }
    public DateTimeOffset Created { get; init; }

    public List<EmpathyInsight> Insights { get; } = [];
    public PointOfView? Pov { get; set; }
    public List<Idea> Ideas { get; } = [];
    public List<Prototype> Prototypes { get; } = [];
    public List<TestResult> Tests { get; } = [];
}

public enum DesignPhase
{
    Empathize,
    Define,
    Ideate,
    Prototype,
    Test
}

public record EmpathyInsight(
    string Observation,
    string UserQuote,
    string Interpretation
);

public record PointOfView(
    string User,
    string Need,
    string Insight
);

public record Idea(
    string Title,
    string Description,
    int Votes,
    string Cluster
);

Related Skills

  • stakeholder-analysis - Identifying interview subjects
  • journey-mapping - Empathize phase visualization
  • business-model-canvas - Modeling solutions
  • prioritization - Selecting which ideas to prototype

Version History

  • v1.0.0 (2025-12-26): Initial release

平台分布

Antigravity

25.34%
按下载量换算386

OpenCode

21.63%
按下载量换算329

Claude Code

17.35%
按下载量换算264

Gemini CLI

13.73%
按下载量换算209

trae

7.71%
按下载量换算117

Cursor

3.49%
按下载量换算53

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。

继续浏览同类 Skills