Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/jiushiwon/wg-skillsnpx agentmods add commands/jiushiwon/wg-skills/auditWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/jiushiwon/wg-skills/audit)<a href="https://agentmods.dev/commands/jiushiwon/wg-skills/audit"><img src="https://agentmods.dev/badge/commands/jiushiwon/wg-skills/audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/jiushiwon/wg-skills/audit"><img src="https://agentmods.dev/badge/commands/jiushiwon/wg-skills/audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01459 |
| Opus 5 | $0.00000 | $0.00730 |
| Sonnet 5 | $0.00000 | $0.00292 |
| Haiku 4.5 | $0.00000 | $0.00146 |
Grade A, and why
audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
audit — 设计 + 代码双维审查
综合审查项目：设计法则、可访问性、AI slop、性能、代码质量。
用法
audit [path] [--depth quick|standard|deep] [--focus design|code|all]
深度
| 深度 | 检查数 | 时间 |
|---|---|---|
quick |
8 项核心 | 1-2 分钟 |
standard |
24 项标准 | 5-10 分钟 |
deep |
60+ 项深度 | 30+ 分钟 |
焦点
| Focus | 说明 |
|---|---|
design |
设计法则 + 可访问性 + AI slop |
code |
代码质量 + 性能 + 安全性 |
all |
全部（默认） |
检查清单
设计维度
调色（4 项）
- 使用 OKLCH 或现代色彩空间
- 中性色向品牌色相微调（chroma 0.005-0.01）
- 无纯
#000/#fff（除非故意） - 调色策略明确（Restrained / Committed / Full / Drenched）
字体（4 项）
- 字号阶 ≥1.25 比例
- 行长 65-75ch（桌面）/ 35-60ch（移动）
- 行高 1.5-1.75（正文）/ 1.1-1.3（标题）
- 无 Inter / Roboto / Arial 等 AI slop 字体
布局（4 项）
- 4pt 网格（间距）
- 圆角在阶梯上
- 触摸目标 ≥44pt
- 嵌套卡片 = 0
动效（3 项）
- 仅 transform / opacity
- 缓动 = ease-out 指数
- prefers-reduced-motion 支持
可访问性（6 项）
- 文本对比度 ≥4.5:1（AA）
- 大文本对比度 ≥3:1
- 焦点环可见
- 键盘可达
- 语义化 HTML
- 减弱动效支持
AI Slop（6 项）
- 无侧边色条边框
- 无渐变文字
- 无玻璃拟态默认
- 无大数字 hero-metric
- 无相同尺寸卡片网格
- 无紫色渐变 + 白底
代码维度
性能（4 项）
- 图片 WebP / AVIF
- 字体预加载 + font-display: swap
- 代码分割 / 懒加载
- 资源懒加载
安全性（3 项）
- 无硬编码密钥
- 输入校验
- XSS 防护
可维护性（3 项）
- TypeScript 严格模式
- ESLint / Prettier
- 组件命名一致
响应式（3 项）
- viewport 正确（不禁用缩放）
- 移动优先
- 安全区处理（iOS）
流程
1. 检测项目
node scripts/detect-stack.mjs
2. 扫描文件
# 收集文件列表
node scripts/extract-tokens.mjs > tokens.json
3. 逐项检查
按上述清单扫描，每项给出 pass / warn / fail / na。
4. 评分
总分 = pass * 1.0 + warn * 0.5 + fail * 0
5. 输出报告
输出：audit-report.md
# Audit Report — <project>
> 时间：2024-XX-XX
> 深度：standard
> 焦点：all
---
## 总分
**78 / 100**（21 pass / 3 warn / 0 fail）
## 设计维度（12/12 项检查）
### 调色
- ✅ 使用 OKLCH
- ✅ 中性色微调
- ✅ 无纯 #000/#fff
- ⚠️ 调色策略未明确
### 字体
- ✅ 字号阶 1.25
- ✅ 行长 70ch
- ✅ 行高 1.6
- ❌ 使用了 Inter
...
## 代码维度（12/12 项检查）
### 性能
- ✅ WebP
- ✅ 字体预加载
...
## 优先级修复（Top 5）
1. **替换 Inter 字体**（AI slop）
- 影响：品牌形象
- 修复：使用 Geist / General Sans 替代
- 工作量：低
2. ...
## 验证
- 构建：✅ pass
- Token 验证：✅ pass
- 减弱动效：✅ pass
- AI slop：❌ fail（Inter）
## 建议
（按优先级排列的修复建议清单）
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 213 lines · 0 tokens per session scan A 4400bf4f272a
audit is a command published in the GitHub repository jiushiwon/wg-skills (97 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,459 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
critique
Design lens critique covering visual hierarchy, clarity, and anti-slop patterns — produces a findings table, no code edits unless asked. Use when the user wants a design review, says "what's wrong with this UI", or needs a second opinion before a handoff or presentation.
hld-review
Review High-Level Design (HLD) against architecture principles and requirements.
pi-ui-review
UI/UX review via Gemini — accessibility, responsive design, component structure.
ui-flow-review
Review menus, HUD, navigation, and player flow from a UX perspective.
design-system-review
Analyze design token usage and component consistency.
verify-design
Carefully and systematically review the provided design document to verify it is consistent, aligned with the current codebase, and free of meaningful gaps. Fix everything you can, then ask about the rest.