For macOS, generically, you can run `screencapture -o -l $WINDOW_ID output.png` to screenshot any window. You can list window IDs belonging to a PID with a few lines of Swift (that any agent will generate). Hook this up together and give it as a tool to your agents.
I think different people mean different things by "see the pixels." I can't imagine being able to see individual pixels on a 15" 4k display at laptop distances, but I can imagine being able to notice minor distortions in the outlines of text, or more discomfort than necessary when reading small text, as a result of lower pixel densities. That could be reasonably described as seeing the pixels.
My iPhone has ~320dpi and I can't see pixels at non-ridculous viewing distances. I don't know what GP's viewing distance habits or eyes are like but for me, considering laptop viewing distance is probably like twice that of smartphones, I can't imagine a 15" 4K laptop being anything but overkill.
Bitcoin is divisible to 0.00000001, so if a bitcoin was worth $10 million, the smallest possible transaction would still be worth just 10 cents (+ transaction fees).