📦 UI Audit — 界面评估
v1.0.1基于 Tommy Geoco《Making UX Decisions》原则，一键扫描界面视觉层级、无障碍、认知负荷、导航等 UX 维度，自动生成可操作建议，助力设计师与开发者快速优化体验。
22· 8.1k·53 当前·54 累计
by @tommygeoco下载技能包
最后更新
2026/2/26
安全扫描
OpenClaw
安全
high confidenceThis is an instruction-only UI audit skill that is internally consistent with its stated purpose: it uses local reference files and checklists, requires no credentials or installs, and does not ask for unrelated system access.
评估建议
This skill appears coherent and low-risk: it only contains local UX guidance files and no credential or install requirements. Before using: (1) verify the author/homepage (audit.uxtools.co / GitHub link in package.json) if provenance matters to you; (2) avoid pasting private or sensitive design URLs/files unless you trust the agent/environment (reports may include sensitive UI content); (3) if you install via npm or a skills registry, prefer installing from the official repo or registry to avoid...详细分析 ▾
✓ 用途与能力
Name/description (UI audit) match the provided contents: SKILL.md and many references are design checklists and frameworks. The package.json, README, and references all align with an auditing/checklist skill. There are no unrelated credentials, binaries, or platform requirements.
✓ 指令范围
Runtime instructions are limited to loading the included reference files, following checklists, and generating audit reports based on user-supplied designs (e.g., Figma URLs). The SKILL.md does not instruct the agent to read system files, environment variables, or to transmit data to unexpected external endpoints.
✓ 安装机制
No install spec and no code files to execute; the skill is instruction-only. package.json exists (for distribution) but there is no install script that pulls remote artifacts or executes code beyond a harmless postinstall echo. Low install risk.
✓ 凭证需求
The skill requests no environment variables, credentials, or config paths. All data the skill would act on is either included (reference files) or provided by the user (design URLs/files), which is appropriate for its purpose.
✓ 持久化与权限
always:false (not force-included). disable-model-invocation:false is the platform default and acceptable here; nothing in the skill requests elevated or persistent privileges or modifies other skills or system settings.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.12026/1/18
README updates: Vercel AI SDK installation instructions
● 无害
安装命令
点击复制官方npx clawhub@latest install ui-audit
镜像加速npx clawhub@latest install ui-audit --registry https://cn.longxiaskill.com