Skip to content

StudyLoop

StudyLoop is a local-first study companion for active learning with an AI mentor. It is designed for learners who benefit from one clear focus, short Socratic questions, visible progress, and a dependable way to return after a break.

StudyLoop Study Session using Kiro
A real StudyLoop session with live Kiro responses; the topic and learner reply were supplied when recording.

Start with one useful session

  1. Follow the Setup Guide to install StudyLoop and connect an agent such as Kiro.
  2. Run studyloop web and open the address printed in the terminal.
  3. Choose Study Session, enter one topic, set your energy, and start.

If you would rather build a routine gradually, Your First Week adds review, plans, and session history one step at a time.

Choose the support you need

Mode Useful when What StudyLoop provides
Study Session You want to understand or practise a concept A live mentor that asks one useful question at a time
Body Double Starting or staying with a task is the hard part A focused workspace, timer, and quiet agent presence
Today Too many possible next steps are competing One recommended action based on time, energy, reviews, and recent work
Flashcards and quizzes Something needs retrieval practice Local review material and spaced-repetition scheduling
Study Plans A larger goal needs shape A plain-language mission, milestones, and checkpoints

Designed with AuDHD learners in mind

Energy-aware pacing, tangible closure, a parking lot for tangents, readable themes, and low-shame recovery are part of the workflow rather than optional polish. Read the AuDHD Learning Philosophy for the reasoning behind those choices.

What stays local

Your StudyLoop database, study plans, generated material, and optional note exports live on your machine by default. The selected AI agent may use its own provider or local model, so its data handling depends on that agent's setup. StudyLoop does not make a local workflow automatically offline: the Web UI needs the local server, and cloud-backed agents still need their provider.

Current boundaries

  • Install from a Git checkout on macOS or Linux; there is no current PyPI or Homebrew release.
  • Use a laptop or tablet for the Web UI. Phone layouts are not supported.
  • Keep studyloop web running while using the browser app; there is no offline service worker.
  • Create study plans through the current Web UI form or CLI. An agent-led planning interview is not integrated into the Web UI yet.
  • Use the CLI for practice-task generation and verification.

See the public roadmap for what is ready, what is being refined, and what is deliberately later.

Find the right guide