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-android-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-android-native)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/software-android-native"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-android-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-android-native"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/software-android-native.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 | $0.00038 | $0.05366 |
| Opus 5.5 | $0.00015 | $0.02146 |
| Sonnet 5.5 | $0.00008 | $0.01073 |
| Haiku 4.5 | $0.00004 | $0.00537 |
Grade A, and why
software-android-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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Native Android Development
Use this skill for native Android work only. It is the default shared-skill entrypoint for Compose-first Android apps targeting API 28+, bounded rewrites from older codebases, and agent-assisted workflows in Android Studio, Codex, and Claude Code.
Quick Reference
| Task | Default Picks | Notes |
|---|---|---|
| State & UI | ||
| New UI screens | Jetpack Compose | Views interop only where existing mature flows or third-party SDKs require it |
| Observable state | ViewModel + StateFlow (Kotlin 2.x) |
Replaces LiveData for new code |
| Async work | Kotlin Coroutines + Flow | Dispatchers.IO for blocking, Dispatchers.Default for CPU; structured concurrency preferred |
| Unit/integration tests | JUnit 4 + Turbine | Android-supported default; JUnit 5 requires verified community integration |
| UI tests | Compose Testing APIs (ComposeTestRule) |
Espresso only for Views interop or legacy screens |
| State machine discipline | ||
| Submit guard | if (_uiState.value is Loading) return |
Prevents double-tap duplicate submissions in ViewModel |
| Transient completion | One-shot UI effect or explicit acknowledgement | Do not reset durable state on an arbitrary timer; cancel or supersede stale work |
| Minimal sealed classes | Remove states that can't happen anymore | Dead sealed subclasses produce dead when branches and mislead future readers |
| Networking & resilience | ||
| Network reachability | ConnectivityManager + NetworkCallback wrapped in StateFlow |
Publish isConnected; disable submit buttons when offline; observe in collectAsStateWithLifecycle |
| DI & architecture | ||
| Dependency injection | Hilt | @HiltViewModel, @Inject constructor, @Module + @InstallIn |
| Local persistence | Room + KSP | Prefer @Upsert over separate insert/update; KSP replaces KAPT |
| Background work | WorkManager + CoroutineWorker |
Deferrable, constraint-aware background processing |
| Runtime tool selection and device proof | software-android-runtime-debugging | Use its tool checks before implementation; build-loop details are in Android Studio workflows |
| Compose, billing, adaptive, auth & push | See Quick Picks | LazyColumn keys, side effects, Play Billing acknowledgement, WindowSizeClass, Credential Manager, FCM, deep links |
collectAsStateWithLifecycle |
stateFlow.collectAsStateWithLifecycle() |
Lifecycle-aware collection; prevents updates when app is backgrounded |
| Strong Skipping (Kotlin 2.x) | ||
| UI state instance identity | Inspect stability/recompositions before splitting state | Unstable parameters use identity, stable ones equality; StateFlow conflates equal values |
LazyListScope lambdas |
Distinguish the non-composable DSL from composable item content | Strong Skipping memoizes lambdas in composable item content; hoist callbacks into the enclosing composable only when a non-composable DSL allocation matters |
| Concurrent state updates | Use MutableStateFlow.update { } for read-modify-write; keep Compose snapshot-state (mutableStateOf) writes on the main thread by convention |
MutableStateFlow.value assignment is itself thread-safe — the real hazard is a non-atomic value = value.copy() race under concurrent writers |
What ships with it
19 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 374 B
- assets/template-android-agent-handoff.md 307 B
- assets/template-android-feature-request.md 743 B
- assets/template-android-proof-checklist.md 799 B
- assets/template-android-rewrite-brief.md 733 B
- data/sources.json 11 KB
- evals/evals.json 7.6 KB
- learnings.consolidated.md 599 B
- learnings.md 863 B
- references/agentic-android-tooling.md 3.6 KB
- references/android-release-and-compliance.md 7.6 KB
- references/android-rewrite-playbook.md 3.6 KB
- references/android-scenarios-and-budgets.md 5.1 KB
- references/android-studio-workflows.md 7.2 KB
- references/codex-claude-android-workflows.md 6.1 KB
- references/compose-state-concurrency.md 23 KB
- references/coroutine-and-compose-recipes.md 7.6 KB
- references/runtime-proof-and-prompts.md 4.2 KB
- references/ui-and-integration-patterns.md 8.8 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 · -131 lines 830562dc6b47
- 1mo ago Changed · +5 lines 9256fead6a90
- 1mo ago First seen · 340 lines · 38 tokens per session scan A 99764796a310
software-android-native is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (101 stars, last pushed 12d ago), licensed MIT. It adds 38 tokens to every session and 5,366 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.