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 vasilyu1983/AI-Agents-public --skill software-ios-nativegit clone --depth 1 https://github.com/vasilyu1983/AI-Agents-publicWrote 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/vasilyu1983/ai-agents-public/software-ios-native)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/software-ios-native"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-ios-native/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/vasilyu1983/ai-agents-public/software-ios-native"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-ios-native.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Rogue Agent · line 35 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Rogue Agent · line 196 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Rogue Agent · line 197 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00039 | $0.07049 |
| Opus 5.5 | $0.00016 | $0.02820 |
| Sonnet 5.5 | $0.00008 | $0.01410 |
| Haiku 4.5 | $0.00004 | $0.00705 |
Grade A, and why
software-ios-native 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 22d 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 — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Native iOS Development
Use this skill for native iOS work only. It is the default shared-skill entrypoint for SwiftUI-first iOS 17+ apps, bounded rewrites from older codebases, and agent-assisted workflows in Xcode, Codex, and Claude Code.
Quick Reference
| Task | Default Picks | Notes |
|---|---|---|
| State & UI | ||
| New UI screens | SwiftUI | UIKit interop only where Apple APIs require it |
| Observable state (iOS 17+) | @Observable + main actor isolation |
Replaces ObservableObject/Published |
| Async work | async/await, structured concurrency |
Detached tasks only when intentionally breaking inheritance |
| Unit/integration tests | Swift Testing | Preferred over XCTest for new tests |
| UI automation tests | XCTest / XCUITest | Keep using for UI and performance tests |
| State machine discipline | ||
| Submit guard | guard state == .idle else { return } |
Prevents double-tap duplicate submissions in @Observable stores |
| Transient completion | One-shot UI effect or explicit acknowledgement | Do not reset durable state on an arbitrary timer; store and cancel any delayed task |
| Minimal state enums | Remove states that can't happen anymore | Dead enum cases produce dead error handling and mislead future readers |
| Networking & resilience | ||
| Network reachability | @Observable singleton + NWPathMonitor |
Publish isConnected; disable submit buttons when offline; start monitor in screen .onAppear |
| Agent tooling & build | ||
| Agent tooling (in Xcode) | Xcode native assistant | Verify the live stable/beta Xcode and Swift versions, and Xcode's macOS host requirement (matters for CI runner images), at developer.apple.com/news/releases before citing one; check Apple's Upcoming Requirements page for the upload-SDK floor and minimum deployment target before submission — do not build release submissions against a beta SDK |
| Agent tooling (outside Xcode) | XcodeBuildMCP if callable | Otherwise fall back immediately to Apple CLI |
| CLI fallback | xcodebuild, simctl, xcresulttool |
Default path when MCP is unavailable or blocked |
| Build / install / stale-app failures | software-ios-runtime-debugging |
Use before UI or feature diagnosis |
| XcodeGen projects | scripts/generate-xcodeproj.sh |
Must regenerate after adding new Swift files |
| Local-dev launcher pair | scripts/run-local-ios-dev.sh + scripts/stop-local-ios-dev.sh |
Two defensive guards (grep env + generated plist for localhost:); persist simulator UDID; never pkill -f Simulator. See quick-reference-extended.md#agent-tooling--build |
.pbxproj-managed projects |
Add new files to target membership | Do not assume on-disk Swift files are auto-discovered |
| Generated files / CI landmines | Commit generated outputs or regenerate in CI hook | git add -f for gitignored-folder tracked files; Xcode Cloud ci_post_clone.sh for env shims. See quick-reference-extended.md |
| TARGETED_DEVICE_FAMILY → "1" | Requires fresh archive + ASC build attachment | Changing project.yml alone is insufficient; value is baked into binary. |
| Canvas, immersive screens, sheets, l10n hygiene | See quick-reference-extended.md | Canvas start state and gestures, persistent sheets, viz state ownership, grid sizing, sheet swapping, l10n plain-enum call rule (catalog parity and regeneration live in software-localisation) |
| StoreKit & billing | ||
| StoreKit 2 subscriptions | @Observable StoreKitManager + TransactionSyncService |
transaction.finish() only after backend sync confirms |
| Paywall presentation | .sheet(isPresented:) from any locked screen |
Don't navigate to Settings; present modal directly |
| Product pricing display | product.displayPrice from StoreKit |
Never hardcode prices; Apple handles locale formatting |
| Promotional offers (rewards) | Server-signed .promotionalOffer() in Product.purchase(options:) |
Bridges Stripe credit gaps for Apple-billed users; user must redeem |
| Paid Apps Agreement (#1 invisible blocker) | Verify Active at appstoreconnect.apple.com/business | Silent empty Product.products(for:) results; check before subscription-level diagnosis. Full decision table + timeline in app-store-connect-checklist.md Phase 5. See quick-reference-extended.md |
ASC Missing Metadata |
Fill all required fields: prices, localization, review screenshot | Products exist but fields incomplete. Workflow in app-store-connect-checklist.md. See quick-reference-extended.md |
| Review screenshot | sRGB 8-bit RGB PNG 72 DPI, 1284 × 2778 | ASC rejects Display P3 / 16-bit device screenshots; use Pillow profile conversion. See quick-reference-extended.md |
| Accounting currency (backend-denominated) | NumberFormatter + explicit currencyCode + locale = .current |
Never String(format: "£%.2f", …); details in quick-reference-extended.md |
| Year/ID display | Text(verbatim:) |
Suppress locale number formatting (2026 not 2,026) |
| Locale-aware time formatting; strings and backend locale | "jm" date skeleton via DateFormatter; strings, plurals and the backend translation pipeline live in software-localisation |
Platform rows in quick-reference-extended.md |
| Auth & push | ||
| Sign in with Apple | ASAuthorizationController + CheckedContinuation |
Guideline 4.8 requires an equivalent privacy-preserving login (limits data to name/email, allows a private relay email, no ad tracking without consent) if a 3rd-party social login is offered — Sign in with Apple is the usual way to satisfy it, not the only one; check current exemptions |
| OTP code input | Hidden TextField + .textContentType(.oneTimeCode) |
Better than magic links for native; iOS auto-fills from notifications |
Non-@MainActor delegates |
Match the SDK requirement; hop to @MainActor for UI state |
MainActor.assumeIsolated checks isolation; use it only when the callback contract guarantees that executor. |
| Push notification categories | Register UNNotificationCategory in didFinishLaunchingWithOptions |
Must be set before any notification arrives; match aps.category from backend |
| Badge count (iOS 16+) | try? await UNUserNotificationCenter.current().setBadgeCount(0) |
applicationIconBadgeNumber is deprecated; setBadgeCount is async throws |
| Push delegate isolation | Inspect the imported SDK requirement and callback contract | Hop UI mutations to @MainActor; do not infer a universal delegate fix from a private crash symbol. See swiftui-observation-concurrency.md and quick-reference-extended.md |
Task { } isolation |
Inherits the actor context where its closure is formed; annotate explicitly across nonisolated callbacks | Task.detached breaks actor inheritance. See swiftui-observation-concurrency.md |
actor vs @MainActor final class |
Choose the isolation owner by the work and shared state | A main-actor caller resumes on the main actor after await; keep CPU-heavy work off it. See swiftui-observation-concurrency.md |
@Sendable async closure awaited from @MainActor |
Return results to the isolated caller; isolate the body if it needs UI state | @Sendable is a transfer-safety constraint, not an executor choice; caller isolation survives await. See swiftui-observation-concurrency.md |
SCNView in UIViewRepresentable |
Create initial scene once; update changed inputs | A no-op updateUIView is correct only for immutable inputs. See swiftui-observation-concurrency.md |
| Push action routing | Check response.actionIdentifier in didReceive |
UNNotificationDefaultActionIdentifier = tap; custom IDs = action buttons; dismiss = no route |
| Push-open ownership, preferences, entitlements, APNs proof, export gate, backend routing, QA loop | See full rows in quick-reference-extended.md | Inspect final exported app/profile: push-enabled TestFlight/App Store exports require production; backend: per-device push_environment column authoritative |
| Swift Concurrency crash triage | Symptom-first triage runbook | Private SwiftUI symbol crash → start at swift-concurrency-crash-triage.md; ladder: console → MTC → TSan → lldb bt |
| SwiftUI API modernization | ||
| Deprecated API review | references/swiftui-deprecated-api.md | Systematic deprecated→modern mapping |
| SwiftUI performance audit | references/swiftui-performance.md | View splitting, lazy stacks, modifier efficiency |
| Modern Swift idioms | references/modern-swift-patterns.md | Foundation modernization, date/string/collection patterns |
| Concurrency (constructive) | ||
| Writing correct concurrency | references/swift-concurrency-patterns.md | Structured concurrency, async streams, bridging, migration |
| Concurrency compiler errors | references/swift-concurrency-diagnostics.md | Error→fix mapping for Swift 6 diagnostics |
| Persistence | ||
| SwiftData modeling | references/swiftdata-core.md | Core rules, predicates, CloudKit, indexing, class inheritance |
| Core Data persistence | references/core-data-persistence.md | Stack setup, contexts, object IDs, batch ops, migrations, CloudKit |
| Reusable app skeleton | references/native-ios-app-foundation-skeleton.md | SwiftUI shell, Observation state, persistence, CloudKit, App Intents, local AI hooks, release gates |
| iCloud database app | references/icloud-cloudkit-app-skeleton.md | SwiftData/Core Data/CloudKit choice, private/public/shared scopes, no-server limits |
| Stacks & monetization | ||
| Pick a starter stack to monetize/engage | references/starter-stacks-and-monetization.md | CloudKit→Cloudflare→RevenueCat→Supabase graduation ladder; on-device AI as free tier; webhook idempotency traps |
| Run iOS dev as a conveyor / app factory | references/ios-app-conveyor.md | 4 pillars: default stack per class, shared SPM, Fastlane+Match CI, agent build loop; 2026 stack survey |
| Build performance | ||
| Xcode build optimization | references/xcode-build-optimization.md | Benchmarking, diagnostic flags, SPM analysis, common wins |
| Swift 6.2+ / Xcode 26+ additions | See quick-reference-extended.md | Default MainActor isolation, iOS 26 withAnimation regression, stale SubscriptionStatus.all, AnyView/@ObservedObject/NavigationView/.id(UUID()) modernization, Combine lifecycle, coordinator navigation |
What ships with it
40 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.
- agents/openai.yaml 352 B
- assets/scaffolds/app-class-blueprints.md 6.1 KB
- assets/scaffolds/cloudflare-worker-backend.md 5.8 KB
- assets/scaffolds/entitlement-and-paywall.md 6.6 KB
- assets/scaffolds/push-and-engagement.md 5.6 KB
- assets/template-ios-agent-handoff.md 271 B
- assets/template-ios-cloudkit-persistence-stack.md 1.9 KB
- assets/template-ios-feature-request.md 728 B
- assets/template-ios-makefile-and-proof-loop.md 1018 B
- assets/template-ios-proof-checklist.md 684 B
- assets/template-ios-rewrite-brief.md 624 B
- assets/template-native-ios-app-skeleton.md 1.5 KB
- data/sources.json 19 KB
- evals/evals.json 12 KB
- learnings.consolidated.md 595 B
- learnings.md 1.5 KB
- references/agentic-ios-tooling.md 3.1 KB
- references/codex-claude-ios-workflows.md 5.7 KB
- references/core-data-persistence.md 3.6 KB
- references/icloud-cloudkit-app-skeleton.md 4.5 KB
- references/ios-app-conveyor.md 11 KB
- references/ios-release-and-compliance.md 5.6 KB
- references/ios-rewrite-playbook.md 2.2 KB
- references/ios-traps-and-scenarios.md 12 KB
- references/modern-swift-patterns.md 8.0 KB
- references/native-ios-app-foundation-skeleton.md 5.9 KB
- references/networking-and-data-loading.md 8.1 KB
- references/quick-reference-extended.md 16 KB
- references/runtime-proof-and-prompts.md 4.8 KB
- references/starter-stacks-and-monetization.md 14 KB
- references/swift-concurrency-diagnostics.md 7.6 KB
- references/swift-concurrency-patterns.md 17 KB
- references/swiftdata-core.md 11 KB
- references/swiftui-composition-patterns.md 5.1 KB
- references/swiftui-deprecated-api.md 11 KB
- references/swiftui-observation-concurrency.md 29 KB
- references/swiftui-performance.md 11 KB
- references/ui-and-integration-patterns.md 4.7 KB
- references/xcode-build-optimization.md 10 KB
- references/xcodebuildmcp-workflows.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.
- 22d ago Changed · -57 lines 77d294130656
- 1mo ago Changed 6b17a66d77f6
- 1mo ago First seen · 309 lines · 39 tokens per session scan A 5d92c02bd98e
software-ios-native is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (101 stars, last pushed 12d ago), licensed MIT. It adds 39 tokens to every session and 7,049 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
kotlin-android
Use when building or fixing a native Android app in Kotlin and Jetpack Compose on the UDF layered architecture — ViewModel/StateFlow, Hilt, Room, Retrofit, coroutines, type-safe Navigation, and the Gradle/AGP surface. NOT shared Android and iOS UI from one Kotlin codebase (that is compose-multiplatform).
compose-multiplatform
Use when building one shared Compose UI in Kotlin across Android, iOS, and desktop — commonMain @Composables, expect/actual, source-set placement, native interop, multiplatform ViewModel/navigation/Koin. NOT a single-platform native build (that is kotlin-android / swift-ios), and NOT Dart/Flutter cross-platform UI…
kotlin-patterns
Kotlin: coroutines, Flow, sealed/data classes, null safety, Ktor, Compose, KMP. Triggers: Kotlin, coroutine, Flow, suspend, Ktor, Jetpack Compose, KMP, kotlinx.
swift-patterns
Swift/iOS: SwiftUI, Combine, async/await, actors, SPM, Core Data, UIKit interop. Triggers: Swift, SwiftUI, Combine, iOS, Xcode, actor, Core Data, @MainActor, @State.
kotlin-android
Build and upgrade native Android applications with Kotlin, Compose/View UI and lifecycle-aware work. Use when implementing Android screens, navigation, permissions or background jobs; coordinate Kotlin, Gradle, AGP and SDK compatibility.
swift-ios
Build and upgrade native iOS applications with Swift, SwiftUI/UIKit and correct lifecycle/concurrency ownership. Use when implementing native screens, platform integrations or Swift migrations; verify toolchain, deployment target and device behavior separately.