返回技能列表

review-ux-ui

pjt222
更新于 Yesterday
4 次查看
17
2
17
在 GitHub 上查看
设计design

关于

This skill performs comprehensive UX/UI reviews by evaluating interfaces against Nielsen's heuristics, WCAG 2.1 accessibility standards, and user flow efficiency. It's designed for pre-release usability checks, accessibility audits, and form/conversion optimization. Developers can use it to analyze cognitive load, keyboard/screen reader compatibility, and overall interaction effectiveness.

快速安装

Claude Code

推荐
主要方式
npx skills add pjt222/agent-almanac -a claude-code
插件命令备选方式
/plugin add https://github.com/pjt222/agent-almanac
Git 克隆备选方式
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/review-ux-ui

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Review UX/UI

Eval UX/UI → usability|access|effectiveness.

Use When

  • Pre-release usability check
  • WCAG 2.1 AA|AAA audit
  • Flow eval → efficiency, err prevention
  • Form review → conversion
  • Heuristic eval existing UI
  • Cognitive load + info arch

In

  • Required: App (URL|prototype|src)
  • Required: Target user (role, skill, ctx)
  • Optional: User research (interview|survey|analytics)
  • Optional: WCAG target (A|AA|AAA)
  • Optional: Flows to eval
  • Optional: Assistive tech (screen reader|switch)

Do

Step 1: Heuristic Eval (Nielsen 10)

#HeuristicQRating
1Status visibilitySys informs user?
2Match real worldFamiliar lang/concepts?
3User control/freedomUndoredo
4ConsistencySimilar elements behave same?
5Err preventionPrevent before occur?
6Recognition > recallOptions visible?
7Flex/efficiencyShortcuts for experts?
8Aesthetic/minimalEvery element serves purpose?
9Err recoveryErrs clear, specific, constructive?
10Help/docsEasy find?

Severity:

SevDesc
0Not problem
1Cosmetic
2Minor
3Major
4Catastrophic — fix pre-release
## Heuristic Evaluation Findings
| # | Heuristic | Severity | Finding | Location |
|---|-----------|----------|---------|----------|
| 1 | System status | 3 | No loading indicator during data fetch — users click repeatedly | Dashboard page |
| 3 | User control | 2 | No undo for item deletion — only a confirmation dialog | Item list |
| 5 | Error prevention | 3 | Date field accepts invalid dates (Feb 30) | Booking form |
| 9 | Error recovery | 4 | Form submission error clears all fields | Registration |

→ All 10 evaluated w/ findings + severity.

If err: time-tight → focus 1, 3, 5, 9.

Step 2: Access Audit (WCAG 2.1)

Perceivable

  • 1.1.1 Non-text: Imgs have alt (decorative=alt="")
  • 1.3.1 Info/rels: Semantic HTML (headings, lists, tables, landmarks)
  • 1.3.2 Sequence: DOM order=visual order
  • 1.4.1 Color: Not sole means
  • 1.4.3 Contrast: ≥4.5:1 normal, ≥3:1 large
  • 1.4.4 Resize: 200% no loss
  • 1.4.11 Non-text contrast: UI ≥3:1
  • 1.4.12 Spacing: line 1.5x, letter 0.12em, word 0.16em

Operable

  • 2.1.1 Keyboard: All operable
  • 2.1.2 No trap: Focus never trapped
  • 2.4.1 Skip links: Avail
  • 2.4.3 Focus order: Logical
  • 2.4.7 Focus visible: Indicator clear
  • 2.4.11 Not obscured: Not behind sticky
  • 2.5.5 Target size: ≥24x24px (44x44 touch)

Understandable

  • 3.1.1 Lang: lang on <html>
  • 3.2.1 On focus: No surprise
  • 3.2.2 On input: No surprise w/o warn
  • 3.3.1 Err id: Clear text
  • 3.3.2 Labels: Visible
  • 3.3.3 Suggest: Fix hint

Robust

  • 4.1.1 Parsing: Valid HTML
  • 4.1.2 Name/role/val: Custom → ARIA
  • 4.1.3 Status msgs: Announced

→ WCAG 2.1 AA checked w/ pass/fail.

If err: auto-tools first (axe-core, Lighthouse) → manual for judgement.

Step 3: Keyboard + Screen Reader

Keyboard

Tab|Shift+Tab|Enter|Space|Arrow|Esc only:

## Keyboard Navigation Audit
| Task | Completable? | Issues |
|------|-------------|--------|
| Navigate to main content | Yes — skip link works | None |
| Open dropdown menu | Yes | Arrow keys don't work within menu |
| Submit a form | Yes | Tab order skips the submit button |
| Close a modal | No | Escape doesn't close, no visible close button in tab order |
| Use date picker | No | Custom date picker not keyboard accessible |

Screen Reader

NVDA|VoiceOver|TalkBack:

## Screen Reader Audit
| Element | Announced As | Expected | Issue |
|---------|-------------|----------|-------|
| Logo link | "link, image" | "Home, link" | Missing alt text on logo |
| Search input | "edit, search" | "Search products, edit" | Missing label association |
| Nav menu | "navigation, main" | Correct | None |
| Error message | (not announced) | "Error: email is required" | Missing live region |
| Loading spinner | (not announced) | "Loading, please wait" | Missing aria-live or role="status" |

→ Full flows tested keyboard + screen reader.

If err: no SR → inspect ARIA + semantic HTML as proxy.

Step 4: Flow Analysis

## User Flow: Complete a Purchase

### Steps
1. Browse products → 2. View product → 3. Add to cart → 4. View cart →
5. Enter shipping → 6. Enter payment → 7. Review order → 8. Confirm

### Assessment
| Step | Friction | Severity | Notes |
|------|---------|----------|-------|
| 1→2 | Low | - | Clear product cards |
| 2→3 | Medium | 2 | "Add to cart" button below the fold on mobile |
| 3→4 | Low | - | Cart icon updates with count |
| 4→5 | High | 3 | Must create account — no guest checkout |
| 5→6 | Low | - | Address autocomplete works well |
| 6→7 | Medium | 2 | Card number field doesn't auto-format |
| 7→8 | Low | - | Clear order summary |

### Flow Efficiency
- **Steps**: 8 (acceptable for e-commerce)
- **Required fields**: 14 (could reduce with address autocomplete + saved payment)
- **Decision points**: 2 (size selection, shipping method)
- **Potential drop-off points**: Step 4→5 (forced account creation)

→ Critical flows mapped, friction rated.

If err: no analytics → assess by step count + complexity.

Step 5: Cognitive Load

  • Density: Per-screen info appropriate?
  • Progressive disclosure: Complex revealed gradual?
  • Chunking: Related grouped (Gestalt)?
  • Recognition > recall: See vs remember?
  • Consistency: Similar tasks → similar patterns?
  • Decision fatigue: Too many choices? (Hick)
  • Working memory: Remember across steps?

→ Load assessed w/ overload|underload areas.

If err: hard objective → squint test (structure visible squinted?).

Step 6: Form Usability

  • Labels: Visible, associated
  • Placeholder: Examples only, not labels
  • Input types: Right (email|tel|number|date) for mobile
  • Validation timing: Blur|submit, not keystroke
  • Err msgs: Specific ("Email needs @") not generic
  • Required: Marked clear
  • Grouping: Related grouped
  • Autocomplete: autocomplete set
  • Tab order: Matches visual
  • Multi-step: Progress indicator
  • Persistence: Data preserved nav-away

→ Each form vs checklist + issues documented.

If err: many forms → prioritize high-traffic (registration|checkout|contact).

Step 7: Write Report

## UX/UI Review Report

### Executive Summary
[2-3 sentences: overall usability, most critical issues, strongest aspects]

### Heuristic Evaluation Summary
| Heuristic | Severity | Key Finding |
|-----------|----------|-------------|
[Summary table from Step 1]

### Accessibility Compliance
- **Target**: WCAG 2.1 AA
- **Status**: [X of Y criteria pass]
- **Critical failures**: [List]

### User Flow Analysis
[Key friction points with severity and recommendations]

### Top 5 Improvements (Prioritised)
1. **[Issue]** — Severity: [N] — [Specific recommendation]
2. ...

### What Works Well
1. [Specific positive observation]
2. ...

→ Prioritized, actionable recs w/ severity.

If err: too many issues → split "must fix" (3-4) vs "should fix" (1-2).

Check

  • All 10 Nielsen rated
  • WCAG checked (min: 1.1.1, 1.4.3, 2.1.1, 2.4.7, 3.3.1, 4.1.2)
  • Keyboard tested key flows
  • SR tested (or ARIA proxy)
  • ≥1 critical flow analyzed
  • Cognitive load assessed
  • Forms evaluated
  • Findings prioritized + actionable

Traps

  • UX vs visual: UX=how works, visual=how looks. Beautiful UI ≠ good UX.
  • Happy path only: Errs, empty, loading, edges = where bugs hide.
  • No real devices: DevTools = proxy. Real catches touch, perf, viewport.
  • Access afterthought: Late = expensive. Early + continuous.
  • Personal pref: "I prefer..." ≠ UX feedback. Cite heuristics|research|patterns.

  • review-web-design — visual review (layout, typo, color)
  • scaffold-nextjs-app — Next.js scaffold
  • setup-tailwind-typescript — Tailwind CSS

GitHub 仓库

pjt222/agent-almanac
路径: i18n/caveman-ultra/skills/review-ux-ui
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

相关推荐技能

executing-plans

设计

该Skill用于当开发者提供完整实施计划时,以受控批次方式执行代码实现。它会先审阅计划并提出疑问,然后分批次执行任务(默认每批3个任务),并在批次间暂停等待审查。关键特性包括分批次执行、内置检查点和架构师审查机制,确保复杂系统实现的可控性。

查看技能

requesting-code-review

设计

该Skill可在完成任务、实现主要功能或合并代码前自动调度代码审查子代理,确保实现符合需求和计划。它支持通过指定git SHA范围进行精准的代码变更审查,帮助开发者在关键节点及时发现潜在问题。核心原则是"早审查、勤审查",适用于开发流程的各个关键阶段。

查看技能

connect-mcp-server

设计

这个Skill指导开发者如何将MCP服务器连接到Claude Code,支持HTTP、stdio和SSE三种传输协议。它涵盖了从安装配置到认证安全的完整流程,适用于集成GitHub、Notion、数据库等外部服务。当开发者需要添加集成、配置外部工具或提及MCP相关功能时,这个Skill能提供实用的操作指南。

查看技能

web-cli-teleport

设计

该Skill帮助开发者根据任务特性选择Claude Code的Web或CLI界面,并指导如何在两种环境间无缝迁移会话。它能分析任务复杂度、迭代需求等要素,推荐最优工作界面和工作流。关键特性包括会话状态管理、环境切换指导和上下文优化建议。

查看技能