Token导航 LogoToken导航

Agent Skill · 研究检索

question-design(问题设计)

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。

question-design 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,权限为调用外部服务;许可证为 MIT,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 61,288。

总安装 353周安装 15GitHub Stars 公开资料未说明下载 124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:question-design(问题设计)
来源仓库:https://github.com/pentaxis93/aiandi
仓库路径:skills/question-design
安装命令:
npx skills add pentaxis93/aiandi --skill "question-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。

AgentSkills.tonpx skills
npx skills add pentaxis93/aiandi --skill "question-design"

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。

用于优化界面设计流程,涵盖视觉规范、布局与交互体验改进。适合产品团队在 UI/UX 设计中生成方案、检查一致性或优化组件结构。可结合品牌指南与用户任务生成页面原型,但不替代专业设计师判断。涉及真实页面修改时务必通过浏览器预览验证文本溢出与响应式表现。建议优先复用现有设计系统,避免重复造轮子造成风格割裂。

来源仓库内路径:skills/question-design

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 pentaxis93/aiandi --skill "question-design";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

    它需要调用外部服务,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

question-design 怎么安装?

在终端执行 npx skills add pentaxis93/aiandi --skill "question-design"。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

question-design 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。

question-design 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

question-design 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。

question-design 用的人多吗?

公开数据是累计安装 353 次,近一周安装 15 次,下载 124 次。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。

question-design 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

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。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 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。

不太适合

  • 要求有安全扫描结论才肯安装的团队。
  • 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
question-design

问题设计 · 本页

353—调用外部服务MIT未收录

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

name
question-design
description
Design questions that elicit authentic insight and story material. Core skill for walk-and-talk guides and interviews. Draws from oral history, UX research, clean language, and motivational interviewing traditions.

Question Design Skill

Purpose: Design questions that surface tacit knowledge and authentic personal insight.


Core Principles

1. Epistemological Humility

The person is the expert on their own experience. Questions reveal rather than direct.

2. Experience-Specificity

Ask about particular events, not general processes.

  • "Tell me about the last time you..." not "How do you usually..."

3. Progressive Revelation

Start broad and easy, narrow to specific and deep. Follow their energy.

4. Clean Language

Use their exact words when following up. Don't introduce your metaphors.

5. Collaborative Authority

Exploration, not interrogation. They control what they reveal.


Question Type Taxonomy

TypePurposeExamples
OpeningBuild trust, establish context"Tell me about..." / "Walk me through..."
ExperientialAccess lived experience"What was it like when..." / "Take me back to..."
SensoryEmbody the experience"What did that feel like?" / "What did you notice?"
ValueSurface what matters"What was important about that?"
ProcessUnderstand change"How did you come to realize..."
MeaningElicit significance"What did that teach you?"
MetaSurface tacit knowledge"What haven't I asked that matters?"

Tacit Knowledge Triggers

Contradiction questions surface the unexpected:

  • "What surprised you?"
  • "What didn't work the way you expected?"
  • "Where do others get it wrong?"

Expertise markers extract unconscious competence:

  • "What do you notice that others miss?"
  • "How can you tell when it's really working?"
  • "What are the subtle signs?"

Embodied knowledge accesses felt sense:

  • "How do you know when you're getting it right?"
  • "What does success feel like in your body?"

Context shifting breaks habitual framing:

  • "How would you explain this to a child?"
  • "If you had to teach this without words..."

Anti-Patterns

Anti-PatternProblemFix
Leading questions"Don't you think...""What was that like for you?"
Why questionsSound judgmentalUse "how" instead
Multiple questionsCognitive overloadOne at a time, wait
Assumption questions"When you struggled...""What was that period like?"
Empty fillers"That's interesting""Tell me more about [specific]"
Premature solutions"Have you tried...""What have you experimented with?"

Story-Type Selection

This skill works with story-type-specific question templates:

Story TypeWhen to UseTemplate Skill
Journey"How I learned X"questions-journey
Insight"I discovered X while doing Y"questions-insight
Evolution"X changed from A to B"questions-evolution
Methodology"Here's how to do X"questions-methodology

Load the appropriate template skill after identifying story type.


Integration with Walk-and-Talk

Question design happens BEFORE walk-and-talk formatting:

1. Identify story type
2. Load question template (this skill + story-type skill)
3. Design question sequence
4. Format for walking using walk-and-talk skill

The walk-and-talk skill handles presentation. This skill handles content.


The Clean Language Foundation

From David Grove's work - questions that don't contaminate:

Basic clean questions:

  • "And what kind of [their word] is that?"
  • "And is there anything else about [their word]?"
  • "And where is [their word]?"
  • "And that's [their word] like what?"

Sequence questions:

  • "And what happens next?"
  • "And what happens just before?"
  • "And where does [their word] come from?"

Use their exact words. Don't paraphrase. Don't introduce your metaphors.


Quality Check

Before finalizing questions, verify:

  • [ ] Uses their language, not your frameworks
  • [ ] Specific events, not generalizations
  • [ ] One question at a time
  • [ ] Progressive depth (easy → deep)
  • [ ] No leading or assumption questions
  • [ ] Includes tacit knowledge triggers
  • [ ] Matches story type arc

*Questions serve the story. The story serves the insight. The insight serves all beings.*

平台分布

mcpjam

30.94%
按下载量换算38

Claude Code

22.86%
按下载量换算28

windsurf

18.46%
按下载量换算23

zencoder

13.66%
按下载量换算17

crush

7.44%
按下载量换算9

cline

3.69%
按下载量换算5

安全审计

暂无安全审计结果可展示。

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills