yeah prompts and skills and roles can encapsulate instruction modules or standard operating procedures that are not obvious in the weights themselves
but if you’re naming your agents or modeling them as “adversaries” you’re mostly just playing house
Dex Horthy says prompts, skills, and roles can encode standard operating procedures that are not visible in model weights. He adds that naming agents or modeling them as adversaries is mostly role-play, responding to @amorriscode's view that super-intelligent systems need no personas.
yeah prompts and skills and roles can encapsulate instruction modules or standard operating procedures that are not obvious in the weights themselves
but if you’re naming your agents or modeling them as “adversaries” you’re mostly just playing house
I don’t want to manage a agents with different personas. If intelligence is truly super there should be no need for roles. Smart generalists rule the world. If you’re making agents with personas, are you actually finding value from it?View quoted post on X
Source: dex · x.comPublished · added here