What Usuba is
Usuba is two parts, and they do different jobs. This page is the map of both.
- Usuba Checker Scripts — the tests. Plain Python, not AI. They load an Omarchy plugin in a real Omarchy environment, run rule checks over its code, and report what is wrong and how to fix it. Available now.
- Usuba Checker Model — the reviewer. A small AI model, trained only on Omarchy plugin data, that runs free on your own laptop. It is built to review a plugin the way an experienced Omarchy reviewer would. In training.
The names
Section titled “The names”| Name | What it is | Status |
|---|---|---|
| Usuba Checker Scripts | The tests that check a plugin | Available now |
| Usuba Standard v0.1 | The rulebook the scripts enforce | Available now |
| Usuba Checker Model | The model that reviews a plugin | In training |
Start with the Checker Scripts
Section titled “Start with the Checker Scripts”They run today, on your own machine, for free.
- Install — get the Checker Scripts onto your machine.
- Run a check — check a plugin and read the output.
- Reading results — rule ids, warnings, and what a failure means.
- Usuba Standard v0.1 — the rulebook behind every result, with a rule id, a level and a fix for every failure. It is the document you check the Checker Scripts against.
The model
Section titled “The model”The Usuba Checker Model is in training. There are no published model results yet. See Model status and Results and methodology.