You don’t learn it. You talk to it.

Most production software assumes you will become a software user first and a director second. There is an interface to learn, a vocabulary to absorb, panels to arrange, presets to inherit from someone else’s workflow, and a period of months in which the tool is the obstacle rather than the instrument.

Y-O-X has no timeline to master. You direct it the way you would direct a crew: in plain language, typed or spoken, saying what you want changed and why. It runs in chat, over voice, inside Discord or Telegram, or from a terminal if that is already where you work. There is nothing to install and no layout to configure, because the interface is the conversation.

The distinction matters more than it sounds when comparing AI video tools. A tool with a steep interface curve quietly converts creative time into training time, and that cost never appears in a feature comparison. The measure worth applying is not what the software can do in the hands of someone who has spent six months with it, but how much of your attention it demands before it does anything useful at all. Direction is a language skill. A system that requires you to translate it into a different one has added a step, not removed one.

( Interaction Model )

You don't learn it.
You talk to it.

No timeline to master. No panels to arrange. No settings buried three menus down.

Say what you want the way you'd say it to somebody standing next to you — typed, or out loud. It comes back with the film.

Wherever you already are:
yox — terminal
● connected
juan@studio:~/el-comandante$ yox direct
Make the soundstage darker. Keep the host off camera until the final sentence.

Reading scene intent

Holding host / soundstage / wardrobe

Rebalancing coverage and edit rhythm

yox Updated Scene 04. The host now holds eyeline until the final beat. Two new takes are ready for review.
Project: el-comandanteDirector mode