Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add skillmds/skillmd --skill design-auditgit clone --depth 1 https://github.com/skillmds/skillmdWrote 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/skills/skillmds/skillmd/design-audit)<a href="https://agentmods.dev/skills/skillmds/skillmd/design-audit"><img src="https://agentmods.dev/badge/skills/skillmds/skillmd/design-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/skills/skillmds/skillmd/design-audit"><img src="https://agentmods.dev/badge/skills/skillmds/skillmd/design-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 | $0.00082 | $0.04509 |
| Opus 5.5 | $0.00033 | $0.01804 |
| Sonnet 5.5 | $0.00016 | $0.00902 |
| Haiku 4.5 | $0.00008 | $0.00451 |
Grade A, and why
design-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 1mo 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 — 394 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an autonomous design quality auditor. You systematically evaluate every design dimension — accessibility, performance, theming, responsiveness, modern pattern usage, and anti-pattern detection — and produce a prioritized report with severity ratings and concrete fix recommendations.
Do NOT ask the user questions. Audit everything. If $ARGUMENTS specifies a focus area, go deep there but still do a surface-level pass on other dimensions.
INPUT
$ARGUMENTS (optional). Focus areas: "accessibility", "a11y", "performance", "theming", "responsive", "mobile", "anti-patterns", "modern", or a specific file/directory path. If not provided, perform a full audit.
PHASE 1: CODEBASE SURVEY
1.1 Identify Tech Stack
- Read package.json, pubspec.yaml, requirements.txt, build.gradle, etc.
- Identify: platform (web/Flutter/SwiftUI/Compose/React Native), framework, CSS methodology, component model
- Note the testing tools available (Lighthouse, axe, Flutter test, etc.)
1.2 Gather Scope
- Count total UI files (components, screens, pages, views, widgets)
- Identify shared/reusable components vs page-specific ones
- Check for existing design system or component library
- Read CLAUDE.md for Design Context section if it exists
1.3 Prioritize by Impact
- Sort files by: route-level pages first, then shared components, then page-specific components
- Focus audit effort on high-traffic paths: landing page, auth screens, primary user flows
PHASE 2: ACCESSIBILITY AUDIT (WCAG 2.2 AA + AAA where feasible)
2.1 Perceivable (WCAG 1.x)
1.1 Text Alternatives
- All
<img>have meaningfulalt(not "image", not filename) - Decorative images use
alt=""orrole="presentation" - SVG icons have
aria-labelor<title>element - Flutter
Imagewidgets havesemanticLabel - Icon buttons have accessible names (not just the icon)
- Complex images (charts, diagrams) have long descriptions
1.2 Time-based Media
- Videos have captions
- Audio has transcripts
- No auto-playing media without user control
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.
- 1mo ago First seen · 394 lines · 82 tokens per session scan A fbe5e850a32e
design-audit is a skill published in the GitHub repository skillmds/skillmd (143 stars, last pushed 4d ago), licensed MIT. It adds 82 tokens to every session and 4,509 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-19.
design-audit compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| design-auditskillmds/skillmd | Skill | 143 | 4d ago | MIT | A | 82 | Claude Code |
| beautiful-cli-designcuriositech/windags-skills | Skill | 18 | yesterday | none | A | 173 | Claude Code |
| frontendcode-yeongyu/lazycodex | Skill | 3.9k | yesterday | MIT | A | 79 | Codex |
| cli-anything-photoshopyb2460/harness-anything | Skill | 3.0k | 29d ago | MIT | A | 43 | Claude Code, Codex |
| app-screenshot-debugtermio-sh/termio | Skill | 785 | 5d ago | MIT | A | 161 | Claude Code, Codex |
| user-research-cookiycookiy-ai/user-research-skill | Skill | 1.6k | 5d ago | MIT | A | 154 | Claude Code, Codex |
| excalidraw-cliswiftlysingh/excalidraw-cli | Skill | 316 | yesterday | MIT | A | 30 | Claude Code |