Show HN: Recall – an MCP server that turns your notes into flashcards

Recall is a Show HN project that runs as an MCP server and turns notes into flashcards. The solo builder made it after repeatedly bouncing off flashcard study because writing cards by hand was too tedious. Connect it to Claude or ChatGPT, point it at handwritten or typed notes—PDFs, docs, or a Notion page—and it writes Q&A cards straight into a deck. You can also skip note-taking: ask the model to research a topic and generate the same cards from that research.

On the product side, Recall sits in the MCP path rather than as a separate study app workflow. An AI client becomes the front end: you attach or reference source material, and the server produces Q&A items and drops them into a deck. Review uses the FSRS spaced repetition algorithm, so scheduling is built in instead of left as a manual export step.

For engineers and builders, the interesting bit is the shape of the tool. Flashcard creation is usually a high-friction, low-automation task; wrapping it as an MCP server means any MCP-capable client can drive ingestion and generation without a custom UI for every source format. That makes PDFs, docs, Notion, and pure research prompts look like the same pipeline: material in, cards out, review on FSRS.

In market terms, this sits where AI assistants and spaced-repetition study tools meet. Most flashcard stacks still expect you to draft cards yourself or use one-off importers; Recall treats card generation as something the assistant does against your notes or a research request. Compared with standalone generators or copy-paste into Anki-style apps, the MCP hook is the differentiator—Claude or ChatGPT is the control surface, and the deck is the sink.

If you already live in Claude or ChatGPT and keep notes in PDFs, docs, or Notion, this is a concrete path from material to a reviewable deck without a hand-written card pipeline. Watch how well multi-source decks stay coherent when cards come from mixed notes and pure research prompts, and whether FSRS review feels complete enough that people stay in Recall rather than exporting elsewhere.