Take any user output — prompt, story, procedure, analysis — and make it maximally effective through iterative intent...
find ~/jbrukh/ -name "*.skill"
Rigorously evaluate whether a prompt or document will produce the expected output when processed by an LLM....