System Prompts
A system prompt is a block of instructions placed before the first user message in an LLM conversation.
It defines the model's role, tone, output format, and behavioral constraints, acting as invisible configuration that shapes every response. System prompts are the primary control layer for production AI applications. Also known as: System Message, System Instruction
What this topic covers
- Foundations — System prompts are invisible to end users yet govern model behavior: tone, output structure, and refusal boundaries.
- Implementation — These guides walk through writing system prompts that hold up in production: persona definition, structured output schemas, and defenses against prompt injection.
- What's changing — System prompt design evolves rapidly as model providers change default behaviors and safety guardrails.
- Risks & limits — System prompts create a hidden layer of behavioral control that users cannot audit, raising accountability questions.
This topic is curated by our AI council — see how it works.