Reusing a brief and building a small library and your first script

The concept

Generation runs through the platform's own app and, on paid tiers, an API; the practice that matters is a saved library of briefs and a consistent naming scheme. The idea is simple: the same model you have been chatting with also answers a web request, so you can put it inside a script, a spreadsheet or a page. A key identifies you; a request sends the text; a response comes back as data.

The reason to try it once, even if you never build anything: it makes the chat version less mysterious. You see that the whole conversation is text in and text out, that your instructions are literally lines of a request, and that "the model" is one parameter among several.

A first call looks like this:

# Name the library by the JOB the track does, not by the prompt.
#
#   assets/music/bed-narration-85bpm.mp3
#   assets/music/bed-narration-85bpm.txt   <- the style line that made it
#   assets/music/jingle-event-120bpm.mp3
#   assets/music/jingle-event-120bpm.txt
#
# The .txt beside each file is the whole trick. In three months you will
# want "the same as the last video" and the prompt box will be empty;
# the style line saved next to the file is what gets it back.
#
# And check the licence for YOUR plan before a track goes on a monetised
# channel or into client work. That is a five-minute check, once.