PeopleInSpace

Kotlin Multiplatform project with SwiftUI, Jetpack Compose, Compose for Wear OS, Compose for Desktop and Compose for Web clients along with Ktor backend. Currently running on

• Android (Jetpack Compose)
• Android App Widget (Compose based Glance API - contributed by https://github.com/yschimke)
• Wear OS (Compose for Wear OS - primarily developed by https://github.com/yschimke)
• iOS (SwiftUI)
• Swift Executable Package
• Desktop (Compose for Desktop)
• Web (Compose for Web - Wasm based)
• Windows (WinUI 3)
• JVM (small Ktor back end service + Main.kt in common module)
• MCP server (using same shared KMP code)

The people data comes from The Space Devs API (names, bios and images of the people currently in space) and the position of the International Space Station from the Where the ISS at? API, both served through this project's own small Ktor backend (see backend module below).

The project is included as sample in the official Kotlin Multiplatform docs and also the Google Dev Library

Architecture

All of the clients are thin UI layers over the same common module: the shared view models expose StateFlows of UI state, backed by a single repository that treats the local SQLDelight database as the source of truth and refreshes it from the network.

flowchart TB
    subgraph clients["Clients"]
        direction LR
        android["<b>app</b><br/>Android · Jetpack Compose<br/>+ Glance / Remote Compose widgets"]
        wear["<b>wearApp</b><br/>Wear OS · Compose + Tile"]
        ios["<b>PeopleInSpaceSwiftUI</b><br/>iOS · SwiftUI (SKIE)<br/>+ SwiftExecutablePackage"]
        desktop["<b>compose-desktop</b><br/>JVM · Compose for Desktop"]
        web["<b>compose-web</b><br/>Kotlin/Wasm · Compose"]
        winui["<b>windows/WinUiApp</b><br/>.NET · WinUI 3"]
        mcp["<b>mcp-server</b><br/>JVM · Kotlin MCP SDK"]
    end

    subgraph common["common (Kotlin Multiplatform)"]
        direction TB
        cmpui["Compose Multiplatform UI<br/><i>PersonList / ISSPosition / ISSMapView</i><br/>(expect/actual map per platform)"]
        vm["View models<br/><i>PersonListViewModel · ISSPositionViewModel</i><br/>StateFlow&lt;UiState&gt;"]
        winclient["PeopleInSpaceClient<br/><i>mingwX64, exported via NuGet</i><br/>(no Koin / AndroidX)"]
        repo["<b>PeopleInSpaceRepository</b><br/>offline-first: DB is source of truth,<br/>ISS position polled every 10s"]
        db[("SQLDelight<br/>PeopleInSpaceDatabase")]
        api["Ktor client APIs<br/><i>PeopleInSpaceApi · AstroviewerApi</i>"]
        koin{{"Koin DI<br/>(annotations + compiler plugin)"}}

        cmpui --> vm
        vm --> repo
        winclient --> repo
        repo --> db
        repo --> api
        koin -.-> repo
    end

    subgraph remote["Remote"]
        direction LR
        backend["<b>backend</b><br/>Ktor / Netty on App Engine<br/><i>/astros.json</i>"]
        spacedevs["The Space Devs API<br/><i>names, bios, images</i>"]
        wheretheiss["wheretheiss.at<br/><i>current ISS position</i>"]
        astroviewer["astroviewer.net<br/><i>predicted ISS orbit</i>"]
    end

    android --> cmpui
    wear --> repo
    ios --> cmpui
    desktop --> cmpui
    web --> cmpui
    winui --> winclient
    mcp --> repo

    api --> backend
    api --> wheretheiss
    api --> astroviewer
    backend --> spacedevs

    classDef client fill:#e3f2fd,stroke:#1565c0,color:#0d1b2a
    classDef shared fill:#ede7f6,stroke:#5e35b1,color:#0d1b2a
    classDef data fill:#e8f5e9,stroke:#2e7d32,color:#0d1b2a
    classDef service fill:#fff3e0,stroke:#ef6c00,color:#0d1b2a
    class android,wear,ios,desktop,web,winui,mcp client
    class cmpui,vm,winclient,koin shared
    class repo,db,api data
    class backend,spacedevs,wheretheiss,astroviewer service

Notes on a few of the edges above:

• The Wear OS client and the MCP server talk to the repository directly (Wear has its own Wear-specific view models, the MCP server just reads the people list).
• The Windows client goes through PeopleInSpaceClient, a self-contained mingwX64 entry point that owns its own Ktor engine, SQLite driver and coroutine scope rather than using Koin.
• Only the people list goes through this project's own Ktor backend; the ISS position and predicted orbit are fetched from their services directly by the shared Ktor client code.

Module overview

Building and running

Requirements: JDK 17, a recent version of Android Studio (for the Android/Wear clients) and Xcode (for the iOS client).

• Android: run the app configuration from Android Studio (or ./gradlew :app:installDebug)

• Wear OS: run the wearApp configuration from Android Studio (or ./gradlew :wearApp:installDebug)

• iOS: open PeopleInSpaceSwiftUI in Xcode and run from there

• Desktop: ./gradlew :compose-desktop:run

• Web (Wasm): ./gradlew :compose-web:wasmJsBrowserDevelopmentRun

• Windows (.NET): see the WinUI 3 client guide for prerequisites and build, test, and run instructions.

• Backend: ./gradlew :backend:run (or run Server.kt directly from Android Studio). After doing that you should then for example be able to open http://localhost:9090/astros_local.json in a browser.

Tests can be run with ./gradlew :common:jvmTest, and there's also a Maestro UI test flow shared by the Android and iOS clients, run by CI on an emulator and a simulator (maestro test -e APP_ID=dev.johnoreilly.peopleinspace maestro/PeopleInSpace.flow).

Backend deployment

Have tested this out in Google App Engine deployment. Using shadowJar plugin to create an "uber" jar and then deploying it as shown below. Should be possible to deploy this jar to other services as well.

./gradlew :backend:shadowJar
gcloud app deploy backend/build/libs/backend-all.jar --appyaml=backend/src/jvmMain/appengine/app.yaml

Screenshots

iOS (SwiftUI + CMP)

Android (Jetpack Compose + CMP)

Wear OS (Wear Compose)

Compose for Desktop

Compose for Web (Wasm based)

MCP

The mcp-server module uses the Kotlin MCP SDK to expose an MCP tools endpoint (returning list of people in space) that can for example be plugged in to Claude Desktop as shown below. That module uses same KMP shared code (that uses for example Ktor, SQLDelight and Koin)

To integrate the MCP server with Claude Desktop you need to firstly run gradle shadowJar task and then select "Edit Config" under Developer Settings and add something like the following (update with your path)

{
  "mcpServers": {
    "kotlin-peopleinspace": {
      "command": "java",
      "args": [
        "-jar",
        "<path-to-repo>/mcp-server/build/libs/serverAll.jar",
        "--stdio"
      ]
    }
  }
}

Related posts

• Minimal Kotlin Multiplatform project using Compose and SwiftUI
• Adding some Storage (to) Space
• Kotlin Multiplatform running on macOS
• PeopleInSpace hits the web with Kotlin/JS and React
• Using Koin in a Kotlin Multiplatform Project
• Jetpack Compose for the Desktop!
• Comparing use of LiveData and StateFlow in a Jetpack Compose project
• Wrapping Kotlin Flow with Swift Combine Publisher in a Kotlin Multiplatform project
• Using Swift Packages in a Kotlin Multiplatform project
• Using Swift's new async/await when invoking Kotlin Multiplatform code
• Exploring new AWS SDK for Kotlin
• Creating a Swift command line app that consumes Kotlin Multiplatform code
• Exploring New Worlds of UI sharing possibilities in PeopleInSpace using Compose Multiplatform
• Moving from SKIE to Swift export in a Kotlin Multiplatform project

Related samples

Note that this repository very much errs on the side of minimalism to help more clearly illustrate key moving parts of a Kotlin Multiplatform project and also to hopefully help someone just starting to explore KMP to get up and running for first time (and is of course primarily focused on use of Jetpack Compose and SwiftUI). If you're at the stage of moving beyond this then I'd definitely recommend checking out KaMPKit from Touchlab. I also have the following samples that demonstrate the use of a variety of Kotlin Multiplatform libraries (and also use Jetpack Compose and SwiftUI).

• GalwayBus (https://github.com/joreilly/GalwayBus)
• Confetti (https://github.com/joreilly/Confetti)
• BikeShare (https://github.com/joreilly/BikeShare)
• FantasyPremierLeague (https://github.com/joreilly/FantasyPremierLeague)
• ClimateTrace (https://github.com/joreilly/ClimateTraceKMP)
• GeminiKMP (https://github.com/joreilly/GeminiKMP)
• MortyComposeKMM (https://github.com/joreilly/MortyComposeKMM)
• StarWars (https://github.com/joreilly/StarWars)
• WordMasterKMP (https://github.com/joreilly/WordMasterKMP)
• Chip-8 (https://github.com/joreilly/chip-8)

Languages, libraries and tools used

• Kotlin
• Kotlin Coroutines
• Kotlinx Serialization
• Ktor client library
• Android Architecture Components
• Koin (using Koin Annotations and the Koin compiler plugin)
• SQLDelight
• Jetpack Compose
• Compose Multiplatform
• Compose for Wear OS
• SwiftUI
• SKIE
• Coil
• Kermit
• Maestro