A prompt is the entire input the model conditions on: the system prompt that sets role and rules, the context pulled in for this request, and the user's words. Engineering it means being explicit about format, showing examples of the desired output, and stating what must not happen — 'answer from the attached price list, and say you do not know when it is not there'.
Its reputation as magic wording is wrong: the durable gains come from structure, not incantations, and a prompt that matters gets tested like code — versioned, eval'd against cases, rolled back when a change regresses.
Related terms
System prompt
The standing instructions sent ahead of every conversation, setting what the model is supposed to do and how it should behave.
Evaluation (evals)
A repeatable test set that measures whether a change to an AI system made it better or worse, rather than just different.
Chain of thought
Prompting a model to reason step by step before answering — accuracy on multi-step problems improves when the working is generated, not skipped.
The bench this belongs to
Generative AIStable Diffusion and ComfyUI wired into the place where your content actually gets made, with a model fine-tuned so everything comes out looking like you.
