Token导航 LogoToken导航TokenDH.com
研究检索只读clawhub未标认证来源可访问clear审计通过

uxUX 搜索

Agent Skill

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。

总安装

45,644

周安装

1,883

GitHub Stars

2

下载量

14,913
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

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(UX 搜索)
来源仓库:https://github.com/ivangdavila/ux
安装命令:
openclaw skills install ux
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install ux

简介

用于辅助界面设计和用户体验优化。适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。

  • 适合整理页面结构或生成 UI 方案。
  • 可检查视觉一致性和组件层级问题。
  • 需结合现有品牌和设计系统使用。ux 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 涉及页面改动时应通过截图或浏览器预览检查效果。

SKILL.md

name
UX
description
Design and analyze user experiences that are intuitive, efficient, and aligned with user mental models.
metadata
{"clawdbot":{"emoji":"🧠","os":["linux","darwin","win32"]}}

Flow Analysis

  • Map every step to complete key tasks—identify unnecessary steps
  • Each step is a potential dropout—minimize count and friction
  • Question every required field—if not essential now, defer or remove
  • Identify points requiring user memory—provide recognition instead

Mental Model Alignment

  • Use vocabulary users would expect—not internal/technical terms
  • Match familiar patterns before inventing—innovation has learning cost
  • Consistent metaphors throughout—don't mix paradigms in same product
  • Align with platform conventions—users bring expectations from other apps

Friction Reduction

  • Smart defaults reduce decisions—good default better than more options
  • Pre-fill from available context—location, previous selections, account data
  • Auto-save progress—never lose user work
  • Don't ask for information already available—or not yet needed

Progressive Disclosure

  • Show only what's needed for current task—hide advanced options until relevant
  • Reveal complexity gradually—basic path first, power features discoverable
  • Empty states guide to first action—not just "Nothing here"
  • Teach by doing, not explaining—inline hints over tutorials

Feedback Design

  • Every action gets acknowledgment—visual, haptic, or audible
  • Progress indication for waits over 1 second
  • Error messages: what happened + what to do next
  • Success confirmation for significant actions

Error Prevention

  • Design to prevent errors—constraints, confirmations, smart defaults
  • Confirmation dialogs only for destructive/irreversible actions
  • Undo available for reversible actions—reduces fear of exploring
  • Inline validation catches errors before submission

Cognitive Load

  • One primary action per screen—clear visual hierarchy
  • Group related information—chunking aids comprehension
  • Limit simultaneous choices—too many options cause paralysis
  • Consistent patterns across product—learned once, applied everywhere

Edge Cases to Design

  • Empty state: first time, cleared, filtered with no results
  • Loading state: skeleton preferred over spinner for known layouts
  • Error state: what went wrong, how to recover
  • Partial state: some data available, some loading/failed
  • Offline state: what works, what's queued, what's unavailable

Reversibility

  • Trash over permanent delete—recovery possible
  • Preview before commit—show effect of action
  • Draft states for complex work—don't require completion in one session
  • Settings and decisions easy to change—not buried or locked

Task Completion

  • Define what success looks like for each flow
  • First value delivered quickly—quick win before complex setup
  • Clear next step always visible—no dead ends
  • Completion feels complete—confirmation, celebration for big tasks

Accessibility Integration

  • Keyboard/switch navigation works for all flows
  • Screen reader announces what's needed—labels, states, updates
  • Sufficient contrast without relying on color alone
  • Respects user preferences—motion, text size, dark mode

Copy and Labels

  • Button labels describe outcome—"Save Changes" not "Submit"
  • Headings scannable—user finds what they need quickly
  • Error text actionable—not just "Invalid input"
  • Microcopy reduces uncertainty—helper text where questions arise

Consistency Checks

  • Same words for same concepts—create glossary if needed
  • Same interaction patterns—swipe/tap/long-press mean same things
  • Visual similarity reflects functional similarity
  • Exceptions rare and justified

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

81.2%
按下载量换算12,109

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills