StudyLoopGitHub

Start small

One useful session is enough for day one.

You do not need to configure every feature. Install StudyLoop, connect one agent you already use, and try one topic that matters to you.

Before you begin

What you need

macOS or Linux

Use a laptop or tablet for the Web UI. Phone layouts are not supported yet.

A little terminal help

Installation is from a Git checkout today. There is no PyPI or Homebrew release yet.

One supported AI agent

Kiro CLI, Codex, or Claude Code are core. OpenCode and pi are preview integrations.

Step 1

Install from the public repository

These commands clone StudyLoop, install its tools and agent definitions, create your local configuration, and check the setup.

Terminal
git clone https://github.com/NetDevAutomate/StudyLoop.git studyloop
cd studyloop
./scripts/install.sh
studyloop setup
studyloop doctor --fix

Stop when the core checks pass

Optional tools can wait. One working agent is enough to learn whether StudyLoop helps you.

Step 2

Start the browser workspace

Terminal
studyloop web
  1. 1

    Open the local address

    Keep the terminal running. StudyLoop needs its local server while the Web UI is open.

  2. 2

    Choose Study Session

    Enter one topic, choose your installed agent, and set your energy from 1 to 10.

  3. 3

    Answer one real question

    The mentor should ask rather than lecture. Think aloud; “I am not sure” is useful information.

  4. 4

    End deliberately

    Use the end control so StudyLoop can keep a summary and leave a clear return point.

StudyLoop Study Session with the topic, mentor terminal, timer, and session status visible
The real Study Session surface. Your installed agent may use a terminal or a structured chat view.
Before sharing anything

Local-first is not the same as fully offline

StudyLoop keeps its database, plans, and generated materials on your machine by default. Your chosen AI agent may still send conversation content to its model provider, whose privacy and billing terms apply.

Next

Pick a workflow that fits today

Study Session is only one route. Body Double and Today solve different kinds of friction.

Compare use cases →