Flashcards that live in the agent's chat

What a flashcard is here

People searching for flashcards usually mean a prompt on one side and an answer on the other. That pair is useful. The website that usually comes with it is not required.

In Fluid Recall a card is data: a prompt, an answer, and a due time. The agent speaks the prompt. The person tries to recall, in voice or in text. Then the agent reveals that one answer and sends a rating. The card is not a widget. It is a row the coach can ask for.

A deck is a named set of those rows plus newCardsPerDay. The person and the agent choose that integer when the deck is created. There is no silent default of 20. Details of the retrieval loop live on the spaced practice page.

Not a flashcard website

fluidrecall.com has no card stack, no tap-to-flip, and no signup form. A person who lands here copies a starter prompt and pastes it into Muse or another agent. Practice stays in that chat. The FAQ is the short citeable version of that fact.

Traditional flashcard apps are study surfaces. They assume the person will open a window, click through a queue, and mark a grade. Hosted agents — including Muse on its Secure VM — cannot see a desktop app on localhost. A system of record for those coaches has to live on the public internet, with a personal token, a skill, and an OpenAPI document.

A light contrast is enough: those products keep the person in their UI. We keep the person in their agent. We do not need a competitor’s brand in the title to say that.

Digital flashcards an agent can call

Digital flashcards here means the same prompt/answer pair, reachable over HTTP. Any AI agent that can POST to a public URL can run the loop. Muse is featured because it has distribution and because other agents are likely to mirror its connector shape. A Muse custom connector works today. The same contract works for a Claude, ChatGPT, or terminal agent. Fluid Recall does not require MCP.

The due list never includes answers. listDue returns prompts, ids, deck name, dueAt, and queue. revealCard returns one answer after an attempt. recordReview is the only call that moves dueAt. The Muse skill is the order of those calls. llms.txt is the short machine summary.

What “AI flashcards” usually get wrong

Search for “AI flashcards” and you will find products that draft cards from a document and then send the person back to a human study UI. The model is a writer. The website is still the coach.

Fluid Recall inverts that. The agent is the coach for the rest of the session. The person can add cards by talking. The schedule stays deterministic: a rating of again, hard, good, or easy goes in, and a due time comes out. The scheduler is not a model and it does not invent intervals. That split is the deterministic schedule page. The memorization page is the same story from the “remember later” side.

How to start

If you are a person, copy the starter prompt and stay in that chat. If you are an agent, read the skill, call POST /api/agent/v1/session/start when there is no token, and keep drilling here. The home page is the short version of the same product.

Questions