Personal project / Early alpha / 0.1.0

Made for students
by students

Lura Study turns your PDFs, notes and scripts into practice. Your own AI agent writes the tasks, the app schedules the reviews and shows how ready you are for the test. Everything stays on your computer — it only talks to the agent or API you connect.

  • Local-first
  • MIT
  • Unsigned alpha
  • No account
  • No telemetry

Lura Study / 127.0.0.1 / Local only

Interface preview / Today
Today / What is due and what is next

Why this exists

I built Lura Study for my own exams. I wanted my notes turned into practice without handing them to another service, so the app keeps the material, the tasks and the progress in one folder on my machine. It is still early — but it is the app I study with.

— Joshua

01 / What it is

  1. 01 / From your own material

    PDFs, Word documents, Markdown and plain text. Every task cites the page or section it came from, so you can check the answer against the source.

  2. 02 / Your agent, not ours

    Connect Claude Code over MCP, or any OpenAI-compatible API. The app talks only to what you set up. There is no Lura account and no Lura cloud.

  3. 03 / Practice that schedules itself

    Recall and apply questions, spaced review, and a readiness view per learning goal — not just a pile of flashcards.

  4. 04 / A real exam simulation

    One sitting, a timer, points and a Swiss 1–6 grade. Afterwards: a per-goal breakdown and a fix list for the days before the test.

  5. 05 / Labs with a real terminal

    Optional Podman or SSH labs run real commands, checks and hints. Use them or ignore them.

02 / How it works

  1. Make a workspace and add your sources.

  2. Review the learning goals it extracts.

  3. Ask your agent for tasks, and check them against the source.

  4. Practice daily, then simulate the test.

03 / Screens

Interface preview / Practice
PRACTICE — the question, the source it came from, your answer.
Interface preview / Material
MATERIAL — sources, learning goals and the task browser.
Interface preview / Progress
PROGRESS — time, streak, accuracy and recall per goal.

04 / Install 0.1.0

All downloads live on GitHub Releases. Each release includes SHA256SUMS.txt and latest.json.

Linux x64

AppImage, rpm or deb. Tested on Fedora 44. The deb is built but not yet tested on Debian or Ubuntu.

  • AppImage
    chmod +x Lura-Study-0.1.0-linux-x64.AppImage
    ./Lura-Study-0.1.0-linux-x64.AppImage
  • RPM
    sudo dnf install ./Lura-Study-0.1.0-linux-x64.rpm
  • DEB
    sudo dpkg -i ./Lura-Study-0.1.0-linux-x64.deb

NOTEOn Ubuntu 24.04 and later, AppArmor rules may block the AppImage; use the deb there.

Windows x64

Windows 10 or 11, x64. The installer runs per user by default; no admin prompt.

  1. Run Lura-Study-0.1.0-win-x64.exe.
  2. If SmartScreen appears: More info → Run anyway.
  3. Uninstall later in Settings → Apps → Installed apps.

macOS arm64

Apple Silicon only in 0.1.0. Intel Macs are not built.

  1. Open Lura-Study-0.1.0-mac-arm64.dmg.
  2. Drag Lura Study into Applications.
  3. First start: allow it in System Settings → Privacy & Security → Open Anyway.

If macOS says the app is damaged:

Terminal
xattr -dr com.apple.quarantine "/Applications/Lura Study.app"

Not signed — on purpose

0.1.0 is not code-signed. Certificates cost money this project does not have. Windows and macOS will warn you; the rpm and deb carry no GPG signature. Every download has a SHA-256, and the full source is public — verify the file or build it yourself.

Requirements

64-bit OS. No Node.js needed — the app brings its own. Optional for labs: Podman (Linux) or an SSH target, and script from util-linux. Settings → Doctor tells you what is missing.

Check a download

  • Linux
    sha256sum -c --ignore-missing SHA256SUMS.txt
  • macOS
    shasum -a 256 -c SHA256SUMS.txt
  • Windows PowerShell
    Get-FileHash .\Lura-Study-0.1.0-win-x64.exe -Algorithm SHA256

05 / Your data One folder. No account.

Where your data folder lives on each system
Linux $XDG_DATA_HOME/lura-studyusually ~/.local/share/lura-study
macOS ~/Library/Application Support/lura-study
Windows %APPDATA%\lura-study

Workspaces, backups, logs and secrets.json (readable only by you) live there. The app listens on 127.0.0.1 and needs a token for every request. There is no telemetry. The update check runs only when you press it, and it only fetches latest.json.

Back up from Settings → Data → Back Up Now.

Listens on
127.0.0.1
Port
7421
Accounts
0
Trackers
0

06 / FAQ

Is it free?

Yes. MIT-licensed, no paid tier, no account. Read it, build it, fork it.

Do I need Claude Code?

No. The app runs without an agent. Tasks can come from an MCP agent such as Claude Code, from an OpenAI-compatible API, or from you.

Does my material get uploaded?

Only to the agent or API you connect. Lura Study has no cloud and sends no telemetry.

Why do Windows and macOS warn me?

Because 0.1.0 is unsigned. Use More info → Run anyway on Windows, or Open Anyway in macOS Privacy & Security. Verify the SHA-256 first if you like.

How do updates work?

Settings → About and updates → Check for updates. It tells you when a newer version exists; you download and install it over the old one. Your data stays.

What can't 0.1.0 do?

Image-only PDFs (no OCR yet), automatic updates, and more than one instance at a time (port 7421). Labs need Podman or SSH.

How do I remove it?

Uninstall the app the normal way for your OS. Your data folder stays until you delete it, so back it up first if you want to keep it.