memento-flashcards

You can create local flashcards with spaced-repetition scheduling by running Python scripts within the CLI workflow, which automatically calculates next-review dates and stores all card data in a local JSON file under ~/.hermes.

Yes, you can turn YouTube transcripts into flashcards by using the quiz generation feature, which fetches transcripts via the youtube-transcript-api package and stores the graded results locally as flashcards.

You can batch import and export flashcard decks in CSV format by using the deck management commands, allowing you to organize cards into portable collections and move data in or out of the local JSON storage.

You do not need an internet connection to review due flashcards offline, as all spaced-repetition data and card collections are stored locally in a JSON file, though the YouTube transcript quiz feature requires connectivity.

Free-text flashcard review grading works entirely offline without external services, as the agent evaluates your answers and schedules the next review directly within the local Python CLI workflow.

The best way to manage flashcard collections locally for durable learning is using this CLI workflow, which handles deck organization, spaced-repetition scheduling, and CSV portability without relying on external services.