StudyLoop: turn your markdown notes into flashcards that can't lie to you

Описание и применение

StudyLoop автоматически преобразует заметки в Markdown в карточки для запоминания, проверяя ответы LLM на наличие цитат из исходного текста, чтобы избежать выдуманных фактов.

Задача

Remembering information from markdown notes is hard; manual flashcard creation is slow and LLM‑generated cards risk inventing false answers.

Результат

StudyLoop produced 32 cards from 14 passages, accepted 30 and rejected 2 that lacked supporting quotes, demonstrating reliable card creation. The system allows offline question answering and minimal sync effort after edits.

Рабочий процесс

• Export API key for card generation
• Run python3 -m studyloop add <notes.md> --deck <name> to split notes and generate cards
• Review generated cards; StudyLoop accepts or rejects based on quote containment and content word overlap
• Study cards using python3 -m studyloop study --deck <name> --limit 10 with SM‑2 scheduling
• Ask questions with python3 -m studyloop ask "<question>" --deck <name> for answers sourced from notes

Доказательства

Flashcards fix the remembering part, but making them by hand takes so long that I never kept it up.

Источник

Sometimes the model invents an answer that sounds right and isn't in the notes, and then you memorize something false.

Источник

StudyLoop kept 30 and rejected 2: rejected quote not in passage 484e52b3fc7bd77d

Источник

ask makes no network call. It ranks passages with Okapi BM25 and answers in your notes' own words

Источник