Skip to main content
AI / LLM PromptingCurated

Guardrail Instruction Writer

For compliance-sensitive deployments where refusals must be consistent.

The prompt
Write guardrail instructions for an assistant that must never: [list forbidden behaviors]. Include a graceful refusal template and one example of a borderline request handled correctly.

Anything in square brackets is meant to be replaced. The builder turns those into fields you can type into.

Tired of re-pasting the same context into every new chat? That is what the Unimatrix MCP server is for.

Browse the rest of the library