Build an AI Prompt System for Your Repeatable Service

If you rewrite the prompt from scratch every time you deliver, quality wobbles and you waste effort. This designs a reusable system around your service: a master system prompt, plus the structure to take client input and produce output that hits your non-negotiable quality rules every run. For anyone delivering the same kind of work repeatedly.

I deliver {SERVICE_NAME} repeatedly for clients and want to build a reusable AI prompt system so every output is consistent and high-quality without starting from scratch each time. My typical input from the client is: {CLIENT_INPUT_FORMAT}. My desired output is: {OUTPUT_FORMAT_AND_QUALITY_STANDARD}. My non-negotiable quality rules are: {QUALITY_RULES}. Design a prompt system for me that includes: (1) a master system prompt that sets the AI's role and constraints, (2) a structured intake template I send to every client to gather inputs, (3) a primary generation prompt that uses the intake data, (4) a self-review prompt the AI uses to check its own output against my quality rules, and (5) an instruction for any human editing step I must always do. Format each component as a labelled code block.