Automation and the imagery pipeline and your first script

The concept

Midjourney generation runs through its own app rather than a general API, so the automation to learn here is the pipeline around it: batch prompts, a naming convention, and an edit pass. 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:

# The pipeline, in plain steps:
#
# 1. Write the prompt skeleton once (subject / medium / light / comp / ratio).
# 2. List the 20 subjects you need in a spreadsheet, one per row.
# 3. Generate in batches, and rename each result as you accept it:
#       projectname-01-cover.png
#       projectname-02-inset.png
# 4. Keep only the accepted ones, and keep the prompt for each in the sheet.
# 5. Edit outside the tool for anything that must be exact: text, numbers,
#    a logo, a specific person's face.
#
# Step 5 is the one people skip. The generator makes images, not layout,
# and every headline belongs in your editor, not in the prompt.