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 fusengine/agents --skill design-iosgit clone --depth 1 https://github.com/fusengine/agentsWrote 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/fusengine/agents/design-ios)<a href="https://agentmods.dev/skills/fusengine/agents/design-ios"><img src="https://agentmods.dev/badge/skills/fusengine/agents/design-ios/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/fusengine/agents/design-ios"><img src="https://agentmods.dev/badge/skills/fusengine/agents/design-ios.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00060 | $0.01705 |
| Opus 5.5 | $0.00024 | $0.00682 |
| Sonnet 5.5 | $0.00012 | $0.00341 |
| Haiku 4.5 | $0.00006 | $0.00170 |
Grade A, and why
design-ios 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 13d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The deliverable is a framed HTML mockup plus a SwiftUI-ready handoff spec (named tokens,
text styles, semantic colors, sizing, states, spacing). This skill never writes SwiftUI —
implementation goes to swift-expert or an Apple-platform developer.
Design for Apple Platforms — Mockup and Handoff, Not Code
When
After design-system tokens exist.
Input
design-system.md— OKLCH palette, typography direction, motion personality.- The screen/flow and the target platform(s). Pick the platform reference first:
| Target | Read |
|---|---|
| iPhone | steps below + references/dynamic-type.md |
| iPad, resizable window, iPhone Duo | + references/ipados-iphone-duo.md |
| Mac | + references/macos.md, references/type-scales-other.md |
| Apple Watch, Apple TV, Vision Pro | + references/watch-tv-vision.md, references/type-scales-other.md |
Steps
- Map tokens to platform roles. Typography → the platform's HIG text styles by name
(iOS/iPadOS Dynamic Type: Body 17 pt … Caption 2 11 pt; macOS fixed styles: Body 13 pt
… Large Title 26 pt; tvOS/watchOS tables) —
dynamic-type.md,type-scales-other.md. Colors → semantic roles, never raw hex —semantic-colors.md. - Choose canvases by size class, not by device: at least compact width, regular width,
and the key screen at AX5 text size. The HIG no longer publishes device point sizes —
take them from Xcode Device Hub and record the source —
viewports.md. - Build the HTML mockup: frame at the recorded point size (1 pt = 1 CSS px),
viewport-fit=cover,env(safe-area-inset-*)for Dynamic Island / home indicator —mockup.md. macOS: a standard system window frame, never custom chrome —macos.md. - Liquid Glass only in the functional layer (tab bars, toolbars, sidebars, key
controls), never in content. Regular variant by default and for text-heavy surfaces;
clear only over rich media, with a 35% dark dimming layer over bright content; accent
color on the primary action's background, sparingly. No published corner radius —
concentric nesting instead —
liquid-glass.md. - Verify targets and contrast: platform control sizes (iOS 44×44 pt, macOS 28×28,
tvOS 66×66, visionOS 60×60, watchOS 44×44) and spacing —
touch-targets.md. Contrast 4.5:1 minimum (3:1 for ≥18 pt or bold), 7:1 target for custom colors and small text, checked in light, dark and Increase Contrast —semantic-colors.md. Sources: https://developer.apple.com/design/human-interface-guidelines/dark-mode and https://developer.apple.com/design/human-interface-guidelines/accessibility. - Sidebar ≤ 2 levels. Deeper hierarchy → a split view with a content list between the sidebar and the detail view, not a third disclosure tier. Consider a tab bar first on iPad. Source: https://developer.apple.com/design/human-interface-guidelines/sidebars.
- Platform pass — apply the platform reference picked above (macOS desktop tinting only under the graphite accent; tvOS 60/80 pt safe area and focus growth; visionOS ornaments for controls only; watchOS ≤3 glyph buttons per row; iPhone Duo controls on the side and the fold reserved region).
- Look at the render — screenshot each canvas (light, dark, AX5) and Read the PNGs
before calling the mockup done; fix what you see, then hand off to
design-review. - Write the handoff spec per
handoff-swiftui.md: named tokens only, text styles by name, semantic colors, sizing (fixed/hug/fill), all states (disabled/pressed/focused, plus pointer style on macOS and focus on tvOS), spacing in pt, canvas sources.
What ships with it
11 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.
- references/dynamic-type.md 3.2 KB
- references/handoff-swiftui.md 1.5 KB
- references/ipados-iphone-duo.md 4.1 KB
- references/liquid-glass.md 3.4 KB
- references/macos.md 3.1 KB
- references/mockup.md 1.1 KB
- references/semantic-colors.md 2.8 KB
- references/touch-targets.md 1.5 KB
- references/type-scales-other.md 3.3 KB
- references/viewports.md 2.2 KB
- references/watch-tv-vision.md 3.5 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.
- 13d ago Changed · -19 lines · +23 tokens per session 7cc5ad592109
- 28d ago Changed 56331682541a
- 1mo ago First seen · 121 lines · 37 tokens per session scan A 4c752117beef
design-ios is a skill published in the GitHub repository fusengine/agents (34 stars, last pushed 13d ago), licensed MIT. It adds 60 tokens to every session and 1,705 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
visual
Define a brand's visual identity — logo usage rules and variations, a color palette with contrast ratios, typography and type scale, and photography, illustration and icon style. Use when the user needs to settle their brand colors, fonts, logo rules or imagery style, or formalize the visual assets they already have.
design-review
Create a Figma Slides deck summarizing a component's design journey — research, contract, options, prototype and test results — with title and summary slides. Use when the user needs to present a component's design work to stakeholders or walk a team through it. Not for feedback on a design; see /coaching:ux or…
form-cro
Optimize a web form for completion — cut fields, choose single or multi-step, fix labels, validation, button copy and trust signals. Use when people start a lead, contact, demo or checkout form but abandon it before submitting. Not for the whole signup-to-trial journey; see /marketing:signup-flow-cro.
a11y-audit
Audit a Figma component or frame for accessibility in one pass — design lint, contrast, color-blindness simulation, focus order — with fixes. Use when the user points at a Figma design and wants a quick accessibility check before it goes to development. Not for a full WCAG-EM audit of a built site or app, or mentoring…
icon-composer-app-icon-workflow
Design, preview, validate, and integrate Apple app icons with Icon Composer. Use when a task mentions Icon Composer, .icon documents, Liquid Glass app icons, ictool preview exports, app icon design for iOS, iPadOS, macOS, or watchOS, or guided Icon Composer GUI work through Computer Use.
export-tokens
Export the design tokens (variables) from a Figma file as CSS custom properties, Tailwind config or JSON, optionally saved to a file. Use when the user wants Figma colors, spacing and type values in code, or gives a Figma URL and asks for its tokens. Not for an inventory of the whole design system; see…