Role prompting means telling the model who to be: a senior editor, a security reviewer, a patient tutor. A role can set vocabulary, tone and the standards the model applies, which is useful for reviewing drafts or explaining at the right level. It does not add knowledge the model lacks, and studies have found that personas rarely make answers to factual questions more accurate.
Roles work best together with a clear task, constraints and an output format. These guides show when personas help and when they do not, how a role fits into a well-structured prompt, and a way of thinking about the model as a simulator that plays whatever characters and world the prompt sets up.
The library lists all guides with filters for topic, level, category and reading time.