A modern, open-source Android news reader built with Jetpack Compose.
Categorised feeds • Offline reading • On-device AI chat • Material You theming

SkyNet Reader started in 2016 as a lightweight Java-based news feed app using Volley, ListView, and the Android Support Library. Over the years, the Android ecosystem evolved dramatically — and so did SkyNet Reader.

In early 2026, the project was rebuilt from the ground up: every line of Java was replaced with idiomatic Kotlin, the UI was migrated from Fragments and XML layouts to a single-Activity Jetpack Compose architecture, and modern libraries (Hilt, Retrofit, Room, Coil, WorkManager) replaced the legacy stack. The result is a clean, testable, and maintainable codebase that follows Clean Architecture principles.

Today, SkyNet Reader serves as both a real-world news reader and a reference project demonstrating current Android best practices.

• Categorised news feeds — World, Entertainment, Business, Technology, and Politics tabs with swipeable horizontal paging
• Offline-first architecture — articles cached in Room; read news without a connection
• Article detail view — in-app browser with reader mode, estimated reading time, and font size controls
• Bookmarks — save articles locally for later; swipe-to-delete with undo
• Full-text search — real-time filtering across cached and remote articles
• On-device AI chat — private, offline conversations powered by MediaPipe's Gemma model
• Dark mode & Material You — dynamic colour theming on Android 12+, manual toggle in settings
• Onboarding flow — first-launch walkthrough built in pure Compose
• Background sync — WorkManager refreshes headlines every 6 hours on Wi-Fi
• Dashboard — at-a-glance stats with article counts, bookmark totals, and recent headlines

The app follows Clean Architecture with MVVM at the presentation layer.

┌───────────────────────────────────────────────────────┐
│              Presentation (Compose + ViewModel)        │
│  Single Activity → NavHost → Composable screens        │
│                 observes StateFlow                      │
└────────────────────┬──────────────────────────────────┘
                     │
┌────────────────────▼──────────────────────────────────┐
│              Domain (Use Cases + Models)                │
│  GetNewsFeedUseCase · SearchNewsUseCase                 │
│  BookmarkArticleUseCase · GetBookmarksUseCase           │
└────────────────────┬──────────────────────────────────┘
                     │
┌────────────────────▼──────────────────────────────────┐
│              Data (Repository + DataSources)            │
│  Retrofit (remote) · Room (local) · DataStore (prefs)  │
└───────────────────────────────────────────────────────┘

• Android Studio Iguana (2024.1) or newer
• JDK 17
• Android SDK 35

# Clone the repository
git clone https://github.com/spkdroid/SkyNet-Reader.git
cd SkyNet-Reader

# Build the debug APK
./gradlew assembleDebug

# Install on a connected device
./gradlew installDebug

The debug APK is written to app/build/outputs/apk/debug/.

The project has Firebase dependencies for analytics and crashlytics. To enable them:

1. Create a project on the Firebase Console
2. Download google-services.json and place it in app/
3. Uncomment apply plugin: 'com.google.gms.google-services' in app/build.gradle

Without this file the app still builds and runs — Firebase features are simply inactive.

The chat screen uses an on-device Gemma 3 model. To enable it:

1. Download gemma-3-1b-it-int4.task (~600 MB)
2. Push to the device:

   adb push gemma-3-1b-it-int4.task /data/data/com.news.skynet.debug/files/

3. Open the AI Chat tab in the app

app/src/main/java/com/news/skynet/
├── AppController.kt                 # @HiltAndroidApp + WorkManager config
├── MainActivity.kt                  # Single-Activity Compose host
├── di/                              # Hilt modules (App, Network, Database)
├── data/
│   ├── remote/                      # Retrofit service, DTOs, remote data source
│   ├── local/                       # Room database, DAOs, entities
│   └── repository/                  # NewsRepository, BookmarkRepository
├── domain/
│   ├── model/                       # NewsArticle, NewsCategory
│   └── usecase/                     # GetNewsFeed, Search, Bookmark use cases
├── ui/
│   ├── compose/
│   │   ├── navigation/              # NavGraph, Screen routes, shared VM
│   │   ├── screens/                 # All Composable screens
│   │   ├── components/              # Reusable composables (NewsArticleCard)
│   │   └── theme/                   # Color, Typography, Theme
│   ├── feed/                        # NewsFeedViewModel, UiState
│   ├── detail/                      # ArticleDetailViewModel
│   ├── dashboard/                   # DashboardViewModel
│   ├── search/                      # SearchViewModel
│   ├── bookmarks/                   # BookmarksViewModel
│   ├── chat/                        # ChatViewModel, ChatMessage
│   └── settings/                    # SettingsViewModel
├── util/                            # NetworkResult sealed class
└── worker/                          # NewsRefreshWorker (periodic sync)

SkyNet Reader fetches news from a custom REST API. The backend, written in PHP, aggregates content from various sources and exposes it as JSON.

The project is developed in phases. See roadmap.md for the detailed plan.

Contributions are welcome! Whether it's a bug fix, new feature, or documentation improvement — feel free to open an issue or submit a pull request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Please make sure your code compiles and passes existing tests before submitting.

Ramkumar Velmurugan Portfolio · GitHub

This project is licensed under the MIT License — see the license.md file for details.

Copyright (c) 2016 Ramkumar Velmurugan

• Apache HTTPComponents: For HTTP request handling.
• CustomWebView: For enhanced web content rendering.
• Volley: For seamless and efficient API communication.