To make the images for the notes herein I use a combination of ChatGPT and Midjourney. I prompt the AI for visual metaphors, then feed my favorite ones into the discord bot along with --ar 24:9
to set the desired aspect ratio. I then convert, optimize and munge the resulting file until I have a string of markdown 
through stitching together wget
, ImageMagick, and jpegoptim
using bash —
#! /bin/bash
if [ -z "$1" ]; then
echo "Provide image URL as argument"
exit 1
fi
filename=$(basename "$1")
extension="${filename##*.}"
wget -O "$filename" "$1"
converted_filename="${filename%.*}.jpg"
convert "$filename" "$converted_filename"
jpegoptim --strip-all "$converted_filename"
echo "" | pbcopy
echo "$converted_filename markdown for jpg embed in clipboard."