mobile-architecture
Skill8 starsChanged 58 days agoMITView on GitHub
Designs and reviews mobile architecture across Flutter, native Android, native iOS, React Native and Kotlin Multiplatform. Covers state ownership, data flow, modules, dependency boundaries, navigation, offline data and KMP shared source sets versus host apps. Use when structuring a feature, choosing state management, refactoring layers or deciding where a real class or screen belongs.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- SWEStash/swe-workflow-skillstheir own version, first seen Nov 8, 2026
- LucSena/how-to-build-softwaretheir own version, first seen Oct 11, 2026
- jcoger/skillstheir own version, first seen Oct 11, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- jcoger/skills★ 1same name
- SWEStash/swe-workflow-skills★ 1same name
- LucSena/how-to-build-software★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository27 other files it gives its agents.
- AGENTS.mdAGENTS.md
- mobile-accessibilitySkill
- mobile-analyticsSkill
- mobile-androidSkill
- mobile-authSkill
- mobile-billingSkill
- mobile-ciSkill
- mobile-code-reviewSkill
- mobile-debugSkill
- mobile-detect-stackSkill
- mobile-firebaseSkill
- mobile-flutterSkill
- mobile-hardwareSkill
- mobile-iosSkill
- mobile-kmpSkill
- mobile-migrationSkill
- mobile-networkingSkill
- mobile-notificationsSkill
- mobile-observabilitySkill
- mobile-offline-syncSkill
- mobile-performanceSkill
- mobile-privacySkill
- mobile-releaseSkill
- mobile-rnSkill
- mobile-securitySkill
- mobile-testSkill
- mobile-ui-uxSkill
The file
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 3 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
- Distinctive: line 24
Read the entry point, navigation setup, dependency wiring, module or package graph and two or three representative features before …
- Distinctive: line 36
Testability as a check: If business logic cannot be tested without a device, it is in the wrong layer.
“Common” is measured: kept by at least 6 in 10 of the 3 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.
Files
Mobile Architecture
What makes a mobile codebase keep up with its team, and how each stack expresses it. This file holds the principles; each platform's patterns, libraries and structure live in its reference.
1. Pick the reference
| Project | Read |
| Flutter | references/flutter.md — structure guides and one guide per state management library |
| Native Android (Kotlin), or the Android side of Kotlin Multiplatform | references/android.md |
| Native iOS (Swift) | references/ios.md |
| React Native or Expo | references/react-native.md |
| Kotlin Multiplatform shared code and target boundaries | references/kotlin-multiplatform.md |
2. Map before judging
Read the entry point, navigation setup, dependency wiring, module or package graph and two or three representative features before proposing changes. A coherent structure that differs from the references is not a finding; inconsistency and coupling are.
3. Principles on every stack
- Layers with one direction of dependency. UI renders state and sends events; logic (view models, blocs, stores, use cases) decides; data (repositories and data sources) fetches and stores. UI never calls HTTP clients, databases or platform SDKs directly.
- Unidirectional data flow. State flows down to the UI; events flow up. The UI does not mutate shared state in place.
- Single source of truth. Each piece of data has one owner — usually a repository — that the rest of the app observes. Duplicated copies drift.
- State ownership. Screen state lives with the screen's logic and survives configuration changes where the platform requires it; app-wide state (session, settings) has one explicit owner; server data lives in a cache or repository, not copied into global stores.
- Impossible states unrepresentable. Model loading, content and error as one sealed or discriminated state instead of independent flags.
- Feature boundaries. Organize by feature once there are several; features depend on shared core code, not on each other's internals.
- Dependencies injected. Construct dependencies at a composition root (DI container or initializers) and pass them in; no service locators reached from deep inside features.
- Offline and errors by design. Decide what works offline, how writes sync and how failures reach the user before building screens.
- Testability as a check. If business logic cannot be tested without a device, it is in the wrong layer.
4. Proportion
Match the structure to the app and team. A domain layer, a DI framework or a multi-module build pays off with size and team count; adding them to a small app is also a finding. Propose incremental steps with the files involved, highest impact and lowest risk first.
Output
## Architecture Overview
## Architecture Issues
## Coupling Problems
## Refactoring Suggestions
## Recommended Architecture
## Action PlanCite path:line for every finding.
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
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.

