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 agentmods add skills/ndisisnd/cook/macosnpx skills add ndisisnd/cook --skill macosgit clone --depth 1 https://github.com/ndisisnd/cookWhat 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 | $0.00075 | $0.01948 |
| Opus 5 | $0.00037 | $0.00974 |
| Sonnet 5 | $0.00015 | $0.00390 |
| Haiku 4.5 | $0.00007 | $0.00195 |
Grade A, and why
macos 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 2d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
macOS Standards
Default load: this file only; pull refs only when needed.
Applies to macOS app targets only — never iOS/iPadOS/watchOS/visionOS, even when Info.plist/entitlements globs match. Swift language rules → standards/swift/; universal rules → standards/global/; secret policy → global/refs/security.md.
Scope: full force for new targets and modules. In existing code, match established conventions; propose migrations (persistence stack, deployment target, architecture slices) as separate explicit tasks, never as a side effect.
P0 — Platform Correctness
Platform Baseline (2026)
- Ship universal 2 (arm64 + x86_64) while the deployment target spans Intel-capable macOS; verify with
lipo -info. arm64-only is a deliberate decision →refs/distribution.md. - Build with the current Xcode/SDK — required for Liquid Glass and current-OS behavior. SDK version ≠ deployment target.
- New apps: minimum deployment macOS 15 default; 14 is the practical floor (loses
@Observable, modern scene APIs). Never bump an existing target as a side effect.
App Structure & Lifecycle
- SwiftUI lifecycle (
@main struct App: App); never@NSApplicationMain, hand-writtenmain.swift, orNSApplication.main()for new app code (non-app bundles legitimately differ). @NSApplicationDelegateAdaptoronly for what SwiftUI can't do; keep it a thin bridge. Scope, Dock-reopen gap, URL schemes →refs/windows-and-scenes.md.- AppKit bridging: SwiftUI-first; wrap only the missing control, never a whole screen; strict concurrency on (AppKit is
@MainActor-only). Gaps + pattern →refs/architecture-and-state.md.
Sandbox & File Access
- Sandbox on by default; narrowest entitlements — a missing entitlement means TCC never even prompts. Detail →
refs/sandbox-and-tcc.md. - User files only via user intent (
NSOpenPanel/fileImporter, drag & drop); persist with security-scoped bookmarks, balancing everystart/stop. - Secrets go in the Keychain — never
UserDefaults, plists, or bundled files.
What ships with it
8 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.
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.
- 2d ago First seen · 120 lines · 75 tokens per session scan A 7ef698d4da55
macos is a skill published in the GitHub repository ndisisnd/cook (2 stars, last pushed 22d ago), licensed MIT. It adds 75 tokens to every session and 1,948 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
build-an-app
Walk through the full process of building a digital health app — from planning to implementation.
spezi-platform-selection
Choose between React Native and Apple-native for a digital health app, set up the matching Spezi starter template, and move existing planning briefs into the new project so the coding agent has full context for implementation.
phone-specs-scraper
Scrape phone specifications from GSM Arena, PhoneDB, and alternative sites. Use when: (1) Comparing smartphone specs, (2) Researching device features, or (3) Building phone comparison tools.
cuelabs-mobile-standard
Build, audit, or standardize CueLABS mobile applications. Use for Flutter architecture, Riverpod, routing, repositories, TESTMODE fakes, flavors, Firebase authentication, design tokens, golden tests, mobile CI, device behavior, or web-to-mobile product parity.
ios-accessibility-validator
Use this skill to audit iOS apps for accessibility compliance with Apple Human Interface Guidelines, WCAG 2.2 standards, and iOS accessibility best practices.
ios-app-store-validator
Use this skill when preparing an iOS app for App Store submission. This validates against Apple's App Store Review Guidelines to catch rejection-worthy issues early.