Auturis

A private beta build, for a small group.

You're getting this link directly from Michael because you're one of a handful of people he'd like real feedback from. This page isn't public — no need to pass it along without checking first.

Download

Windows Zip, 321 MB — tested on real hardware, including live use with an API key.
Download
Mac (Apple Silicon) Zip, 251 MB — M-series only for now; Intel Macs aren't supported yet.
Download

Getting started

  1. Download and unzip. On Windows, run TCA-KG.exe from inside the unzipped folder — keep it together with the _internal folder next to it. On Mac, drag TCA-KG.app out of the unzipped folder and double-click it.
  2. Windows will likely show a blue "Windows protected your PC" SmartScreen warning, since this isn't signed with a paid certificate yet. Click More info, then Run anyway.
  3. Mac will likely refuse to open it the first time and say it "cannot be verified" or "is damaged" — same reason, no paid Apple certificate yet. Right-click (or Control-click) the app and choose Open, then confirm in the dialog that appears. If that doesn't work, go to System Settings → Privacy & Security and click Open Anyway next to the message about this app.
  4. You'll be asked for your own API key on first launch (BYOK — bring your own key). Michael will send you one to use.
  5. Give it a real question or decision you're actually facing — that's where it gets interesting.

What it actually looks like

The Auturis entry screen, offering two starting questions: What should I do? and What is right?

Two questions, two kinds of reasoning

You start from one of two places: a decision you're personally facing, or a rule you're weighing whether to impose on other people. The system treats those as genuinely different problems — it doesn't just run the same check twice.

A real conversation in the app, where the assistant asks a follow-up question about a stated maxim instead of jumping straight to a verdict

It asks before it answers

This is a real exchange, not a mockup. Rather than handing you a verdict on the first message, it asks what's actually going on — the same way a careful person would before saying something is or isn't okay.

A botanical tree visualization representing a personal graph of commitments and values, built up over use

Your own reasoning, mapped

Everything you've worked through gets woven into a personal structure — shown here as a tree — that later questions get checked against. It's not a static rulebook; it's built from what you've actually committed to.