This mobile app offers a seamless way to explore and retrieve collectibles information from the video game Control now enhanced with a Gemini-powered question-answering system for an interactive experience.

Data scraped using Python Scrapy library and stored in Firebase Realtime database. Flutter frontend for mobile application. Gemini Question Answering System added

Folder structure

• controlscrape: Scrapy code
• control_lore: Flutter application

Setup for mobile app

Prerequistes - Flutter SDK. How to Install
Clone the gemini_qa branch and cd into control_lore folder to run the following commands.

flutter pub get 
flutter run 

To use the Gemini API, you'll need an API key. If you don't already have one, create a key in Google AI Studio: https://aistudio.google.com/app/apikey.

When running the app, include your API key using the --dart-define flag:

flutter run --dart-define=API_KEY=$GEMINI_API_KEY

If you use VSCode, you can specify --dart-define variables in your launch.json file.

If you use Android Studio or IntelliJ you can use run / debug configurations (https://www.jetbrains.com/help/idea/run-debug-configuration.html).

To Do

• Scrape multimedia youtube links and add to mobile app as url redirect button
• Add local json load and parsing instead of firebase dependency.
• QA System hosted on server(Provided I get resources to host the models lol.)

Images