agentleFS
harshsinghmpharshsinghmp/muse-skills/skills/agency-delivery/mobile/SKILL.md

mobile

Skill1 starsChanged 17 days agoMITView on GitHub

Full mobile app department: iOS (SwiftUI), Android (Compose), cross-platform (React Native/Expo, Flutter), Universal Expo SDK 52, progressive web apps, and app store optimization — routed through seven modes. Use when asked to build, refactor, or ship mobile apps, build universal Expo apps with New Architecture and NativeWind, convert web apps to native/PWA, implement native features (camera, push, offline), or improve app store visibility. Not for websites (webdev) or UI motion (animate mobile refs).

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every invocation resolves to exactly one mode.

    line 38
  2. Distinctive:

    Only the resolved mode's reference is loaded — the rest stay on disk, saving tokens on every run.

    line 50

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

📱 mobile — Mobile Apps Department

One head skill for mobile. Choose the honest path: web already exists → wrap (Capacitor/PWA); React team → Expo; fresh + performance-critical or platform-deep → native. Never pick a stack by hype. Every mode ends with the platform's build running on a device or emulator, not just compiled.


Modes

Every invocation resolves to exactly one mode. Match the request, then load only the matched reference:

ModeTrigger phrasesBehaviorReference
ios"ios app", "swiftui", "app store submission"SwiftUI-first iOS build, HIG compliance, store submissionreferences/ios.md
android"android app", "jetpack compose", "play store"Compose-first Android build, Material 3, Play submissionreferences/android.md
cross"react native", "expo app", "flutter app", "cross-platform mobile"Expo (React Native) or Flutter cross-platform buildreferences/cross.md
pwa"pwa", "make it installable", "offline web app"Installable, offline-capable progressive web appreferences/pwa.md
aso"app store optimization", "aso", "store ranking", "more app installs"App Store Optimization: keywords, listing, ratings loop, review responsesreferences/aso.md
expo"expo", "expo sdk 52", "expo app", "expo router", "nativewind"Universal Expo SDK 52 mobile app: file-based routing, New Architecture, NativeWind v4, deep linksreferences/expo.md
audit"audit the app", "store audit", "launch audit", "compliance audit", "accessibility audit"App Store audit (keywords, listing, ratings, screenshots, compliance check) + mobile a11y audit (labels/roles/hints, touch targets, screen-reader flow)references/audit.md

Only the resolved mode's reference is loaded — the rest stay on disk, saving tokens on every run.


When to Use

  • Building a new mobile app (native, cross-platform, or web-wrapped).
  • Converting an existing web app to iOS/APK or a PWA.
  • Implementing native capabilities: camera, push, biometrics, offline storage.
  • Refactoring or optimizing an existing mobile app.
  • App store optimization: listings, keywords, screenshots, review handling.

Anti-Triggers

  • Websites and web apps → webdev.
  • Scaffolding the project → new-project (it handles Expo/Capacitor presets).
  • Animation inside mobile UI → animate (its mobile reference), not this skill.

Quick Reference

Stack path ladder (decide before any mode)

SituationPath
Working web app; app must mirror itWrap: Capacitor (web stack) or PWA if install/offline only
React/JS team; app is CRUD + contentExpo (React Native)
Fresh start, either team; fast iteration both platformsFlutter
Platform-deep (widgets, watch, AR, fine perf) or single platformNative: SwiftUI / Compose

Decision inputs: team's existing skills > app depth requirements > budget > long-term maintenance.

Verification gate (every mode)

  • Builds on the platform toolchain (xcodebuild, gradle, expo/eas, flutter build).
  • Runs on a device or emulator — captured logs/screenshots as evidence.
  • Store pre-flight passes (bundle size, permissions justified, signing).

Suite contracts

  • Project scaffold with Expo/Capacitor wiring → new-project first.
  • Server/API work the app consumes → webdev (backend mode).

Procedure

  1. Intake. establish the honest stack path (wrap existing web / cross / native) before writing a line — from team skills, app depth, and budget; the answer drives everything.
  2. Resolve the mode. Match the request against the mode table; exactly one mode. Ambiguous → ask one question, then proceed.
  3. Execute the mode playbook. Load references/<mode>.md and follow its Intake → Deliverable → Procedure → Quality gate in order.
  4. Gate and deliver. Pass this file's Verification checklist plus the mode's quality gate, then deliver the artifact where the client expects it.

Pitfalls

  • Choosing a stack by hype instead of the ladder — the expensive mistake.
  • Testing only in the simulator — device performance and permissions differ.
  • Offline as an afterthought — mobile networks are the default environment, not the edge case.
  • Ignoring platform HIGs (Apple/Material) — rejection risk + clunky feel.
  • Push permissions asked at launch with no context — denied forever after.
  • Bundle bloat from web habits — unused polyfills and images ship to every user.
  • ASO treated as a launch-day task — it shapes name, keywords, and screenshots from day one.

Verification

  • Stack path decided via the ladder and stated with rationale before build.
  • The platform's own build toolchain ran green.
  • App runs on a device or emulator with captured evidence.
  • Native permissions justified (only those used, with purpose strings).
  • Offline/degraded-network behavior tested, not assumed.
  • Store pre-flight checks pass (size, signing, listing assets).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.