Long AI Prompting Experiment

Experiment testing the "added value" of long context-dense prompts (versus casual converastional prompts) and the value of a voice "prompt optimisation" agent