UXD Design Assistant

v1.0.0

UXD设计流程助手,专注于用户体验设计流程辅助。当用户需要以下场景时激活:(1) 备战UXDA大赛或学习用户体验设计/服务设计流程 (2) 不知道设计下一步该做什么,需要流程指引 (3) 需要用户研究指导(访谈法、问卷法、观察法、同理心地图等)(4) 需要竞品分析辅助(功能拆解、体验对比、分析报告框架)(5) 需...

0· 50·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for kingsunzhang-openclaw/uxd-design-assistant.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "UXD Design Assistant" (kingsunzhang-openclaw/uxd-design-assistant) from ClawHub.
Skill page: https://clawhub.ai/kingsunzhang-openclaw/uxd-design-assistant
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install uxd-design-assistant

ClawHub CLI

Package manager switcher

npx clawhub@latest install uxd-design-assistant
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The skill name, description, SKILL.md, templates, and three Python helper scripts align: interview outline, competitor analysis, and video script generation — all expected for a UX design assistant. Minor incongruities in repo metadata/README: the README installation example references a different git URL (kingsunzhang-openclaw/UXDA.git) and the README text both states MIT license and then asserts a commercial-use prohibition; that is a legal/metadata inconsistency but does not indicate malicious behavior.
Instruction Scope
SKILL.md instructs the agent to guide users through UX process and, when needed, run local Python scripts (scripts/*) to generate Markdown/CSV/SVG. The runtime instructions do not direct the agent to read arbitrary system files, access environment variables, or transmit data to external endpoints. The scripts accept user-supplied output paths and write local files, which is expected behavior.
Install Mechanism
There is no install spec in the skill manifest (instruction-only), and included code files are small, readable Python scripts with no downloads or external package installation. The README suggests using an external OpenClaw CLI (npm) or git clone for manual installation — normal documentation but not part of the skill manifest. Overall install risk is low.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. The scripts do not read OS environment variables or secret files; they only use command-line arguments and write output to user-specified files. This is proportionate to the stated functionality.
Persistence & Privilege
The skill is not always-enabled (always: false) and is user-invocable. It does not request persistent elevated privileges, does not modify other skills or system-wide agent settings, and contains no autonomous background components. Standard and appropriate for this kind of helper.
Assessment
This skill appears coherent and low-risk: it provides templates and three small Python generators that run locally and produce Markdown/CSV/text/SVG. Before installing or running: 1) Verify the GitHub homepage and repository authenticity (the README references a different clone URL and the license text contains a contradiction — clarify whether commercial restrictions actually apply). 2) Inspect the scripts (they are small and readable) and run them in a sandbox or non-critical directory to avoid accidental overwriting of files when using the --output option. 3) Confirm you have Python 3 on the host; the skill does not require secrets or network access. If you need legal certainty about reuse, request clarification from the author about the license/usage restriction in the README.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

🎨 Clawdis
latestvk97fky5141cpd3c51gtx858q4x85jz23
50downloads
0stars
1versions
Updated 1d ago
v1.0.0
MIT-0

UXD 设计流程助手

授人以渔,而非授人以鱼。

定位

本 Skill 是教学辅助工具,通过方法论指导、框架模板和自动化脚本,帮助参赛师生按正确流程完成用户体验设计和服务设计学习与备赛。不代劳,只引导。

设计流程五阶段

用户进入任意阶段,提供该阶段相应的工具和指引:

调研阶段 → 分析阶段 → 设计阶段 → 测试阶段 → 输出阶段

阶段判断

用户描述当前状态后,判断其所属阶段并告知:

  • 尚未开始 → 调研阶段
  • 有调研数据 → 分析阶段
  • 有分析结论 → 设计阶段
  • 有原型 → 测试阶段
  • 测试迭代完成 → 输出阶段

各阶段核心指导

1. 调研阶段

目标:理解用户、场景、需求

方法论指导

  • 访谈法:人数建议5-8人,半结构化访谈,时长30-60分钟
  • 问卷法:问卷设计→样本收集→定量分析,注意问卷长度控制在15题以内
  • 观察法:现场观察、影子跟踪法(shadowing),记录行为流
  • 文献研究:行业报告、学术论文、竞品公开资料

输出物框架

  • 用户画像(Persona)
  • 用户旅程图(User Journey Map)
  • 使用场景描述
  • 需求池(Need List)

2. 分析阶段

目标:从数据中提炼洞察

方法论指导

  • 亲和图法(Affinity Diagram):将大量定性数据归类分组
  • 同理心地图(Empathy Map):从"说/做/想/感"四维度理解用户
  • 用户故事(User Story):As a... I want to... So that...
  • 服务蓝图(Service Blueprint):前端触点+后端流程+支持系统
  • POA问题分析(Problem-Opportunity-Area)

洞察提炼原则

  • 避免主观假设,每条洞察需有调研数据支撑
  • 优先级排序:Impact × Feasibility 矩阵
  • 聚焦高频痛点和高价值机会点

3. 设计阶段

目标:将洞察转化为解决方案

低/中保真原型

  • 先纸笔原型(Paper Prototype),再数字原型
  • 使用 Figma / Sketch / Adobe XD / Pixso 等工具
  • 低保真原则:只用方框和文字,不做视觉美化
  • 参考:references/wireframe-guide.md

交互逻辑

  • 用户任务流(Task Flow)
  • 页面流程图(Page Flow)
  • 信息架构图(Sitemap)

创新方法

  • HMW(How Might We)问题重构
  • 奥斯本检核表法
  • 概念草图(Concept Sketch)

4. 测试阶段

目标:验证设计假设

可用性测试

  • 测试人数:5人即可发现85%的可用性问题
  • 任务测试法:给用户具体任务,观察完成率/错误率/时间
  • 出声思考法(Think Aloud):让用户边操作边说出想法
  • 参考:references/usability-testing.md

评估指标

  • SUS(System Usability Scale):10题量表,68分为平均分
  • SEQ(Single Ease Question):"完成这个任务有多容易?" 1-7分
  • NPS(Net Promoter Score):推荐意愿

5. 输出阶段

设计文档框架: 参考 references/design-report-template.md

作品集展示原则

  • 讲好故事:背景→研究→洞察→设计→验证
  • 可视化:用图说话,减少大段文字
  • 突出设计决策依据("因为...所以...")

自动化脚本

以下脚本提供结构化输出,减少重复劳动:

脚本命令输出
访谈大纲生成器python3 scripts/interview_generator.py --goal <主题> --count <数量>Markdown访谈大纲
竞品分析表格python3 scripts/competitor_analysis.py --products <产品1,产品2>CSV格式功能对比表
低保真原型模板参考 assets/wireframe-template.svgSVG线框图模板
视频脚本生成器python3 scripts/video_script_generator.py --duration <秒数> --scene <场景描述>分镜脚本

使用原则

  1. 不代劳:只提供方法论、框架、模板;具体内容由用户产出
  2. 循序渐进:根据用户当前阶段提供相应工具,不跳阶段
  3. 鼓励思考:多提问少给答案,引导理解"为什么"
  4. 证据驱动:设计决策需有调研或测试数据支撑

团队协作建议

推荐工具:

  • 文档协作:腾讯文档 / Notion / 语雀
  • 设计协作:Figma(支持多人实时编辑)
  • 进度管理:飞书周报 / 企业微信任务

参考文件

Comments

Loading comments...