The current implementation just injects the original words into the words.json file and continues to use this file for future operations. This makes things easier only for sustaining the original sets of words but does not help with exporting and editing the new ones