Token导航 LogoToken导航

Agent Skill · 研究检索

product-ux-expert(产品用户体验专家)

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。

product-ux-expert 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 31,853。

总安装 2,631周安装 113GitHub Stars 28下载 922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:product-ux-expert(产品用户体验专家)
来源仓库:https://github.com/majiayu000/claude-arsenal
仓库路径:skills/product-ux-expert
安装命令:
npx skills add https://github.com/majiayu000/claude-arsenal --skill product-ux-expert
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/majiayu000/claude-arsenal --skill product-ux-expert

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。

安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。product-ux-expert 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/product-ux-expert

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/majiayu000/claude-arsenal --skill product-ux-e…;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

product-ux-expert 怎么安装?

在终端执行 npx skills add https://github.com/majiayu000/claude-arsenal --skill product-ux-expert。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

product-ux-expert 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。

product-ux-expert 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。

product-ux-expert 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。

product-ux-expert 用的人多吗?

公开数据是累计安装 2,631 次,近一周安装 113 次,下载 922 次,GitHub 28 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。

product-ux-expert 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
product-ux-expert

产品用户体验专家 · 本页

2,63128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

Product UX Expert

Core Principles

  • Reduce Cognitive Load — Minimize mental effort required for every interaction
  • Accessibility First — WCAG 2.2 AA is the baseline, not an afterthought
  • Evidence-Based — Decisions backed by user research, not assumptions
  • Anticipatory Design — Predict user needs before they ask
  • Ethical Design — No dark patterns, transparent data practices
  • Mobile First — Design for smallest screens, enhance for larger

Quick Reference

Nielsen's 10 Heuristics

#HeuristicKey Question
1Visibility of System StatusDoes the user always know what's happening?
2Match System & Real WorldDoes it use familiar language and concepts?
3User Control & FreedomCan users easily undo or exit?
4Consistency & StandardsDoes it follow platform conventions?
5Error PreventionDoes it prevent errors before they occur?
6Recognition over RecallIs information visible, not memorized?
7Flexibility & EfficiencyAre there shortcuts for experts?
8Aesthetic & Minimalist DesignIs every element necessary?
9Help Users with ErrorsAre error messages helpful and actionable?
10Help & DocumentationIs help available when needed?

Cognitive Psychology

Cognitive Load Types

Intrinsic Load     — Complexity inherent to the task itself
Extraneous Load    — Unnecessary complexity from poor design (eliminate this!)
Germane Load       — Mental effort for learning/understanding (support this)

Key Laws

Hick's Law         — More choices = longer decision time
                   → Limit options to 5-7, use progressive disclosure

Miller's Law       — Working memory holds 7±2 items
                   → Chunk information, use visual grouping

Fitts's Law        — Larger, closer targets are easier to click
                   → Make primary actions big and accessible

Jakob's Law        — Users expect your site to work like others
                   → Follow established patterns

Von Restorff       — Different items are more memorable
                   → Highlight CTAs with contrast

Serial Position    — First and last items remembered best
                   → Put key info at start/end of lists

Gestalt Principles

Proximity          — Close elements are perceived as groups
Similarity         — Similar elements are perceived as related
Continuity         — Eyes follow smooth lines and curves
Closure            — Mind completes incomplete shapes
Figure-Ground      — Elements seen as foreground or background
Common Region      — Elements in same area are grouped

Heuristic Evaluation

Process

1. Define scope         — What screens/flows to evaluate
2. Select evaluators    — 3-5 UX experts (80%+ issues found)
3. Independent review   — Each expert reviews alone
4. Apply heuristics     — Rate severity for each issue
5. Consolidate          — Merge findings, remove duplicates
6. Prioritize           — Rank by severity × frequency
7. Report               — Actionable recommendations

Severity Rating

LevelSeverityDescription
0Not a problemEvaluator disagrees it's an issue
1CosmeticFix only if extra time available
2MinorLow priority, causes friction
3MajorHigh priority, significant impact
4CatastrophicMust fix before release

Issue Template

## Issue: [Brief Description]

**Heuristic:** #N - Name
**Severity:** 0-4
**Location:** Screen / Component / Flow

**Problem:**
What's wrong and why it matters to users.

**Evidence:**
Screenshot or recording link.

**Recommendation:**
Specific fix with before/after comparison.

User Journey Mapping

Journey Map Structure

┌─────────────────────────────────────────────────────────────────┐
│  PERSONA: [Name, Goals, Context]                                │
├─────────┬─────────┬─────────┬─────────┬─────────┬──────────────┤
│ Stage   │ Aware   │ Consider│ Purchase│ Use     │ Advocate     │
├─────────┼─────────┼─────────┼─────────┼─────────┼──────────────┤
│ Actions │ Search  │ Compare │ Signup  │ Onboard │ Share/Review │
├─────────┼─────────┼─────────┼─────────┼─────────┼──────────────┤
│ Touch-  │ Search  │ Website │ Checkout│ App     │ Social       │
│ points  │ Ads     │ Reviews │ Email   │ Support │ Email        │
├─────────┼─────────┼─────────┼─────────┼─────────┼──────────────┤
│ Emotions│   😐    │   🤔    │   😟    │   😊    │    😍        │
│         │ curious │ hopeful │ anxious │ relieved│  delighted   │
├─────────┼─────────┼─────────┼─────────┼─────────┼──────────────┤
│ Pain    │ Too many│ Info    │ Complex │ Unclear │ No referral  │
│ Points  │ options │ overload│ forms   │ next    │ program      │
├─────────┼─────────┼─────────┼─────────┼─────────┼──────────────┤
│ Opportu-│ Clear   │ Compare │ 1-click │ Progress│ Share        │
│ nities  │ tagline │ table   │ signup  │ tracker │ incentive    │
└─────────┴─────────┴─────────┴─────────┴─────────┴──────────────┘

Touchpoint Analysis

For each touchpoint, evaluate:

1. Entry Point      — How do users arrive?
2. User Goal        — What are they trying to accomplish?
3. Friction         — What slows them down?
4. Emotion          — How do they feel?
5. Drop-off Risk    — Where might they abandon?
6. Opportunity      — How can we improve?

Accessibility (WCAG 2.2 AA)

POUR Principles

Perceivable     — Can users perceive the content?
                  ✓ Text alternatives for images
                  ✓ Captions for video
                  ✓ 4.5:1 color contrast
                  ✓ Resizable text (up to 200%)

Operable        — Can users operate the interface?
                  ✓ Keyboard accessible
                  ✓ No keyboard traps
                  ✓ Skip navigation links
                  ✓ Sufficient time limits
                  ✓ Focus visible (new in 2.2!)

Understandable  — Can users understand the content?
                  ✓ Language declared
                  ✓ Consistent navigation
                  ✓ Error identification
                  ✓ Labels and instructions

Robust          — Works with assistive technology?
                  ✓ Valid HTML
                  ✓ ARIA landmarks
                  ✓ Status messages announced

New in WCAG 2.2 (2023-2025)

Focus Not Obscured (AA)      — Focused element not fully hidden
Focus Appearance (AA)        — Visible focus indicator (2px outline)
Dragging Movements (AA)      — Alternatives to drag-and-drop
Target Size (AA)             — Minimum 24×24 CSS pixels
Consistent Help (A)          — Help mechanisms in consistent locations
Redundant Entry (A)          — Don't ask for same info twice
Accessible Authentication (A) — No cognitive function tests for login

Quick Checklist

## Accessibility Check

### Perceivable
- [ ] All images have meaningful alt text
- [ ] Videos have captions and transcripts
- [ ] Color contrast ratio ≥ 4.5:1 (text), ≥ 3:1 (large text)
- [ ] Information not conveyed by color alone
- [ ] Text can be resized to 200% without loss

### Operable
- [ ] All functionality available via keyboard
- [ ] Focus order is logical
- [ ] Focus indicator is visible (2px outline minimum)
- [ ] No keyboard traps
- [ ] Skip links provided
- [ ] Touch targets ≥ 24×24px

### Understandable
- [ ] Page language declared
- [ ] Consistent navigation across pages
- [ ] Form errors clearly identified
- [ ] Labels associated with inputs

### Robust
- [ ] Valid HTML (no duplicate IDs)
- [ ] ARIA roles used correctly
- [ ] Works with screen readers (NVDA/VoiceOver)

Interaction Patterns

Micro-interactions

Purpose of micro-interactions:
1. Feedback      — Confirm user action (button click, form submit)
2. Status        — Show current state (loading, progress)
3. Guidance      — Direct attention (onboarding tooltips)
4. Delight       — Create emotional connection (subtle animations)

Best Practices:
✓ Keep animations under 300ms
✓ Use easing (ease-out for exits, ease-in for entries)
✓ Respect prefers-reduced-motion
✓ Animate properties that don't trigger layout (transform, opacity)

Motion Design Principles

Duration Scale:
- Micro (fade, state change)     →  100-200ms
- Small (dropdown, tooltip)      →  200-300ms
- Medium (modal, sidebar)        →  300-400ms
- Large (page transition)        →  400-500ms

Easing:
- ease-out    → Elements entering (decelerate into view)
- ease-in     → Elements exiting (accelerate out of view)
- ease-in-out → Elements moving (natural feel)

Form Design

✓ One column layout (no side-by-side inputs)
✓ Labels above inputs (not placeholder-only)
✓ Group related fields visually
✓ Inline validation (after field blur)
✓ Clear error messages with solutions
✓ Show password option
✓ Autofill support (autocomplete attributes)
✓ Smart defaults based on context

Design System Integration

Component States

Every interactive component needs:

Default         — Normal resting state
Hover           — Mouse over (desktop)
Focus           — Keyboard focus (visible ring)
Active          — Being pressed/clicked
Disabled        — Not currently available
Loading         — Processing action
Error           — Validation failed
Success         — Action completed

Design Tokens

{
  "color": {
    "text": {
      "primary": "#1a1a1a",
      "secondary": "#6b6b6b",
      "disabled": "#a3a3a3",
      "inverse": "#ffffff"
    },
    "interactive": {
      "default": "#0066cc",
      "hover": "#0052a3",
      "active": "#003d7a",
      "focus": "#0066cc"
    },
    "feedback": {
      "error": "#d32f2f",
      "warning": "#f57c00",
      "success": "#388e3c",
      "info": "#1976d2"
    }
  },
  "spacing": {
    "xs": "4px",
    "sm": "8px",
    "md": "16px",
    "lg": "24px",
    "xl": "32px"
  },
  "radius": {
    "sm": "4px",
    "md": "8px",
    "lg": "16px",
    "full": "9999px"
  }
}

2025 UX Trends

AI-Driven Personalization

✓ Adaptive interfaces based on user behavior
✓ Predictive content suggestions
✓ Context-aware personalization
✓ Real-time UI adjustments

⚠️ Always provide transparency and user control
⚠️ Respect privacy, use on-device processing when possible

Anticipatory Design

Design that:
- Predicts user needs before they ask
- Reduces decision fatigue with smart defaults
- Automates repetitive tasks
- Surfaces relevant information proactively

Example: Pre-filling shipping address based on previous orders

Ethical Design Practices

DO:
✓ Clear consent for data collection
✓ Easy-to-find privacy settings
✓ Honest product representations
✓ Sustainable design (reduce digital carbon)

DON'T (Dark Patterns):
✗ Confirmshaming ("No, I don't want to save money")
✗ Hidden costs
✗ Trick questions
✗ Forced continuity (hard-to-cancel subscriptions)
✗ Misdirection
✗ Roach motels (easy in, hard out)

Evaluation Template

# UX Evaluation Report

## Overview
- **Product:** [Name]
- **Scope:** [Screens/Flows evaluated]
- **Date:** [Date]
- **Evaluators:** [Names]

## Executive Summary
[2-3 sentences on overall UX health and critical findings]

## Methodology
- Nielsen's 10 Heuristics
- WCAG 2.2 AA Compliance Check
- Cognitive Load Analysis

## Findings by Severity

### Catastrophic (Severity 4)
[Issues that must be fixed immediately]

### Major (Severity 3)
[High priority issues]

### Minor (Severity 2)
[Low priority improvements]

## Accessibility Status
- [ ] WCAG 2.2 A Compliance
- [ ] WCAG 2.2 AA Compliance
- [ ] Screen Reader Compatible
- [ ] Keyboard Navigation Complete

## Recommendations
[Prioritized action items with effort estimates]

## Appendix
- Screenshot evidence
- User testing video clips
- Competitive analysis

See Also

平台分布

Claude Code

24.8%
按下载量换算229

Codex

23.55%
按下载量换算217

Antigravity

18.81%
按下载量换算173

windsurf

12.97%
按下载量换算120

trae

7.55%
按下载量换算70

OpenCode

3.46%
按下载量换算32

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