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 robzilla1738/roberts-skills --skill macos-designgit clone --depth 1 https://github.com/robzilla1738/roberts-skillsWrote 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/robzilla1738/roberts-skills/macos-design)<a href="https://agentmods.dev/skills/robzilla1738/roberts-skills/macos-design"><img src="https://agentmods.dev/badge/skills/robzilla1738/roberts-skills/macos-design/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/robzilla1738/roberts-skills/macos-design"><img src="https://agentmods.dev/badge/skills/robzilla1738/roberts-skills/macos-design.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 | $0.00075 | $0.02152 |
| Opus 5 | $0.00037 | $0.01076 |
| Sonnet 5 | $0.00015 | $0.00430 |
| Haiku 4.5 | $0.00007 | $0.00215 |
Grade A, and why
macos-design 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
macOS Design
Hub skill for designing, critiquing, and scaffolding native macOS applications. Read this file first, then open only the spoke files relevant to the task.
Mission
Create macOS applications that feel unmistakably native: powerful, spacious, precise, keyboard-friendly, accessible, and visually calm. Favor standard macOS structures and system components before custom UI. When custom UI is justified, make it harmonize with the system rather than compete with it.
This skill is written for an LLM. Use it as a design-generation and design-review checklist. It is not a substitute for Apple's Human Interface Guidelines, current SDK documentation, or testing in the latest Xcode/macOS, but it is structured to make generated macOS applications substantially more native and polished.
Non-negotiable design stance
A generated macOS app should assume the following unless the user explicitly asks otherwise:
- Use native app structure. A Mac app is not a web page in a resizable window. It has a menu bar, app menu, keyboard shortcuts, window management, settings, undo/redo where appropriate, contextual menus, help, accessibility, and standard system behaviors.
- Respect the content. The interface should frame, organize, and accelerate the user's work. Avoid decorative chrome, gratuitous backgrounds, and UI effects that compete with content.
- Prefer system components. SwiftUI and AppKit controls automatically adapt to appearance, accent color, accessibility settings, localization, input modes, and future OS updates. Custom controls inherit none of this unless you build it.
- Design for density and precision. Mac users expect fast scanning, accurate selection, keyboard access, multiwindow workflows, drag and drop, resize behavior, and visible structure.
- Use Liquid Glass selectively. In the current macOS design language, Liquid Glass belongs primarily to top-level controls, navigation, floating bars, and system-like chrome. Do not turn every card, table row, inspector, or content panel into glass.
- Treat the menu bar as part of the app. Commands belong in predictable menus. Context menus and toolbars are accelerators, not replacements for the menu bar.
- Honor safe areas and the camera housing. Never place essential content or controls where the menu bar, toolbar, rounded display corners, or camera housing can obscure them.
- Accessibility is part of the design, not a pass at the end. Every generated design should include keyboard, VoiceOver, contrast, Reduce Transparency, Increase Contrast, Reduce Motion, and localization considerations.
- Target OS matters. A design using macOS 26-era APIs should include availability checks and graceful fallbacks for earlier macOS versions.
- Do not imitate macOS superficially in standard app windows. Fake traffic lights, fake title bars, fake menu bars, fake notches inside normal window content, and web-style side navigation usually make apps feel less native. Dedicated notch-style utility apps are a separate case — see the macos-notch skill (separate plugin).
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- accessibility.md 3.6 KB
- appkit-patterns.md 6.4 KB
- critique-checklists.md 8.7 KB
- foundations.md 14 KB
- icons-and-visual-language.md 6.2 KB
- layout-and-windowing.md 17 KB
- liquid-glass.md 5.9 KB
- macos_immersive_onboarding_guide.md 11 KB
- navigation.md 4.4 KB
- notch-effect.md 1.0 KB
- README.md 4.1 KB
- swiftui-patterns.md 11 KB
- toolbars-and-menus.md 10 KB
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 · 128 lines · 75 tokens per session scan A e3a898f63b73
macos-design is a skill published in the GitHub repository robzilla1738/roberts-skills (5 stars, last pushed 3mo ago), licensed MIT. It adds 75 tokens to every session and 2,152 once invoked, about $0.0004 per session on Opus 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 2026-08-31.
Other skills, from other repositories
hatch3r-design-system-detect
Detects existing design tokens, component library, and theming convention in a project before authoring new UI primitives — output a concise inventory for downstream implementers.
ss-studio
Turn a product brief and optional references into three distinct creative directions, a human-selected StyleSeed interaction plan, generated image/video asset jobs, a working UI prototype, and a verified prototype-first showcase reel. Use for client concepts, app interaction exploration, trendy but coherent UI…
ss-copy
Generate UX microcopy (button labels, error messages, empty states, toasts) following a casual-but-polite voice and tone.
conversational-ux
Design voice and conversational interfaces — dialog flows, error recovery, and persona. Use when the interface speaks and listens rather than being tapped. For graphical input collection, use form-design.
form-design
Design a form end to end — field order, grouping, validation, and completion. Use when the artifact is a form. For product-wide error strategy use error-handling-ux; for first-run signup use onboarding-design.
peak-end-rule
Apply the Peak-End Rule — a flow is remembered by its most intense moment and its last. Use when designing completion, celebration, or cancellation moments. For sustaining engagement mid-flow, use zeigarnik-effect.