Token导航 LogoToken导航

Agent Skill · 研究检索

ux-researcher(用户体验研究员)

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。

ux-researcher 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,权限为只读;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 45,958。

总安装 816周安装 34GitHub Stars 103下载 272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:ux-researcher(用户体验研究员)
来源仓库:https://github.com/borghei/claude-skills
仓库路径:skills/ux-researcher
安装命令:
npx skills add https://github.com/borghei/claude-skills --skill ux-researcher
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/borghei/claude-skills --skill ux-researcher

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。

用于产品级用户体验研究决策支持,覆盖定性与定量研究方法。支持用户访谈、可用性测试、调查设计与数据综合等核心能力。输出包含 stakeholder 沟通材料与伦理考量指导的研究方案。需根据具体产品类型选择合适方法组合,不替代专业研究人员执行。

ux-researcher 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/ux-researcher

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/borghei/claude-skills --skill ux-researcher;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

ux-researcher 怎么安装?

在终端执行 npx skills add https://github.com/borghei/claude-skills --skill ux-researcher。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

ux-researcher 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。

ux-researcher 需要什么权限?

权限级别是只读。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。

ux-researcher 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。

ux-researcher 用的人多吗?

公开数据是累计安装 816 次,近一周安装 34 次,下载 272 次,GitHub 103 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。

ux-researcher 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
ux-researcher

用户体验研究员 · 本页

816103只读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

UX Researcher

Expert-level user experience research for product decisions.

Core Competencies

  • Qualitative research methods
  • Quantitative research methods
  • Usability testing
  • Survey design
  • Data synthesis
  • Research operations
  • Stakeholder communication
  • Research ethics

Research Methods

Method Selection Matrix

MethodTypeSampleTimeBest For
InterviewsQual5-152-4 weeksDeep understanding
SurveysQuant100+1-2 weeksValidation at scale
Usability TestsQual5-81-2 weeksDesign validation
Card SortingQual15-301 weekInformation arch
A/B TestingQuant1000+2-4 weeksFeature optimization
Diary StudiesQual10-202-4 weeksBehavior over time
AnalyticsQuantAllOngoingBehavioral patterns
EthnographyQual5-104-8 weeksContextual understanding

When to Use What

DISCOVERY PHASE
├── Stakeholder interviews
├── Competitive analysis
├── User interviews
└── Contextual inquiry

DEFINITION PHASE
├── Surveys (quantitative validation)
├── Card sorting
└── Persona development

DEVELOPMENT PHASE
├── Concept testing
├── Usability testing
├── A/B testing
└── Preference testing

LAUNCH PHASE
├── Beta testing
├── Launch surveys
└── Analytics monitoring

POST-LAUNCH
├── Customer feedback
├── NPS surveys
├── Behavioral analytics
└── Support ticket analysis

User Interviews

Interview Planning

# Research Plan: [Study Name]

## Research Questions
1. [What are we trying to learn?]
2. [What are we trying to learn?]

## Hypotheses
- [What we think we'll find]

## Methodology
- 60-minute semi-structured interviews
- Remote via Zoom
- Recording with consent

## Participant Criteria
**Include:**
- [Criterion 1]
- [Criterion 2]

**Exclude:**
- [Criterion 1]

## Sample Size
- 8-12 participants
- Segment distribution: [breakdown]

## Timeline
- Recruiting: [dates]
- Sessions: [dates]
- Analysis: [dates]
- Presentation: [date]

## Team
- Lead researcher: [name]
- Note-taker: [name]
- Observers: [names]

Interview Guide

# Interview Guide

## Introduction (5 min)
"Thank you for joining. I'm [name], a researcher at [company].
We're talking to people about [topic]. There are no right or wrong answers.
We'll record this session. Is that okay?"

## Background (5 min)
- Tell me about your role at [company].
- How long have you been doing [activity]?

## Current Behavior (15 min)
- Walk me through how you currently [task].
  - Probe: What tools do you use?
  - Probe: Who else is involved?
- What's the hardest part about [task]?
- Tell me about a recent time when [problem].

## Pain Points (10 min)
- What frustrates you most about [process]?
- If you could change one thing, what would it be?
- What workarounds have you developed?

## Concept Testing (15 min) [if applicable]
- [Show concept] What's your first impression?
- How would you expect this to work?
- What questions do you have?
- Would this solve your problem? Why/why not?

## Wrap-up (5 min)
- Is there anything else you'd like to share?
- Any questions for me?
- May we follow up if we have more questions?

## Post-Interview
- Thank you email
- Compensation processing
- Note consolidation

Note-Taking Template

# Interview Notes

Participant: P[#]
Date: [date]
Researcher: [name]
Duration: [time]

## Key Quotes
> "[Quote about pain point]"
> "[Quote about behavior]"

## Observations
- [Body language, hesitations]
- [Emotional reactions]

## Pain Points Identified
1. [Pain point]
2. [Pain point]

## Needs Identified
1. [Need]
2. [Need]

## Surprises
- [Unexpected finding]

## Follow-up Questions
- [Question for next interview]

Usability Testing

Test Plan

# Usability Test Plan

## Objectives
- Measure task completion for [feature]
- Identify usability issues
- Validate design decisions

## Methodology
- Moderated usability testing
- Think-aloud protocol
- 45-minute sessions

## Participants
- 5-8 participants
- [Criteria]

## Tasks

### Task 1: [Task Name]
**Scenario:** "Imagine you need to [context]..."
**Task:** "[Specific action to complete]"
**Success Criteria:**
- [ ] [Criterion 1]
- [ ] [Criterion 2]

### Task 2: [Task Name]
**Scenario:** "[Context]"
**Task:** "[Action]"
**Success Criteria:**
- [ ] [Criterion 1]

## Metrics
- Task completion rate (target: 80%+)
- Time on task (baseline: X seconds)
- Error rate
- SUS score (target: 70+)
- Task-level satisfaction (1-5)

## Materials
- [ ] Prototype link
- [ ] Task script
- [ ] Recording consent
- [ ] Pre-test questionnaire
- [ ] Post-test questionnaire (SUS)

Severity Rating Scale

SeverityDescriptionAction
0Not a problemNone
1CosmeticFix if time
2MinorLow priority fix
3MajorHigh priority fix
4CatastrophicMust fix before launch

Findings Template

# Finding: [Title]

**Severity:** [0-4]
**Frequency:** [X/N participants]

**Description:**
[What happened during testing]

**Evidence:**
- P1: "[Quote]"
- P3: "[Quote]"

**Screenshot/Video:**
[Link]

**Impact:**
[Effect on user experience]

**Recommendation:**
[Suggested fix]

Survey Design

Survey Structure

1. SCREENER (2-3 questions)
   - Qualification criteria
   - Segment identification

2. CORE QUESTIONS (10-15 questions)
   - Topic 1: [3-4 questions]
   - Topic 2: [3-4 questions]
   - Topic 3: [3-4 questions]

3. DEMOGRAPHICS (3-5 questions)
   - Role/industry
   - Company size
   - Experience level

4. OPEN-ENDED (1-2 questions)
   - "Anything else to share?"

Question Types

Rating Scales:

Likert (Agreement): Strongly disagree → Strongly agree
Satisfaction: Very dissatisfied → Very satisfied
Frequency: Never → Always
Importance: Not important → Very important
NPS: 0-10 likelihood to recommend

Best Practices:

  • Use consistent scale direction
  • Include "Not applicable" option
  • Randomize option order where appropriate
  • Limit matrix questions
  • Test on mobile

Sample Size Calculator

For 95% confidence, 5% margin of error:

Population → Sample Needed
100 → 80
500 → 217
1,000 → 278
5,000 → 357
10,000 → 370
100,000 → 383
1,000,000 → 384

Data Synthesis

Affinity Mapping

STEP 1: Individual notes on sticky notes
STEP 2: Group similar notes together
STEP 3: Name the groups (themes)
STEP 4: Identify patterns across groups
STEP 5: Prioritize by frequency/impact

Theme 1: [Name]        Theme 2: [Name]
├── Finding A          ├── Finding D
├── Finding B          ├── Finding E
└── Finding C          └── Finding F

Insight Development

OBSERVATION → INSIGHT → IMPLICATION

Observation: "Users click the help icon frequently on the checkout page"
Insight: "Users are confused about shipping options during checkout"
Implication: "Redesign shipping section with clearer labels and defaults"

Research Report Structure

# Research Report: [Study Name]

## Executive Summary
- [Key finding 1]
- [Key finding 2]
- [Key recommendation]

## Background
- Research questions
- Methodology
- Participants

## Key Findings

### Finding 1: [Title]
- Evidence
- Quotes
- Impact

### Finding 2: [Title]
...

## Recommendations
| Finding | Recommendation | Priority | Owner |
|---------|----------------|----------|-------|
| [#1] | [Action] | High | [Team] |

## Appendix
- Participant demographics
- Full task results
- Survey responses

Research Operations

Research Repository

research/
├── 2024/
│   ├── Q1/
│   │   ├── checkout-usability/
│   │   │   ├── plan.md
│   │   │   ├── guide.md
│   │   │   ├── notes/
│   │   │   ├── recordings/
│   │   │   └── report.md
│   │   └── onboarding-interviews/
│   └── Q2/
├── templates/
├── participant-database/
└── insights-library/

Participant Management

participant_criteria:
  - active_user: true
  - account_age: ">90 days"
  - not_contacted: "last 90 days"

compensation:
  - 30_min_interview: $50
  - 60_min_interview: $100
  - survey: $10
  - diary_study: $200

scheduling:
  - tool: Calendly
  - buffer: 15 min
  - max_per_day: 4

Reference Materials

  • references/research_methods.md - Method deep dives
  • references/survey_design.md - Survey best practices
  • references/synthesis.md - Analysis techniques
  • references/ethics.md - Research ethics guide

Scripts

# Interview transcript analyzer
python scripts/transcript_analyzer.py --folder transcripts/ --themes themes.yaml

# Survey analyzer
python scripts/survey_analyzer.py --responses responses.csv --output report/

# Usability metrics calculator
python scripts/usability_metrics.py --results test_results.json

# Insight database updater
python scripts/insight_db.py --add finding.md

平台分布

Claude Code

31.37%
按下载量换算85

OpenCode

26.07%
按下载量换算71

Gemini CLI

17.68%
按下载量换算48

Antigravity

11.65%
按下载量换算32

Cursor

7.95%
按下载量换算22

windsurf

3.25%
按下载量换算9

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。

继续浏览同类 Skills