The 5 Principles at a Glance
Chapter Goal: By the end of Chapter, students will apply all 5 core principles of prompting deliberately, diagnose which principle is missing from any underperforming prompt, and combine principles in a multi-step clinical prompt chain.
P# | Principle | What It Does |
P1 | Give Direction | Tell the AI WHO to be — role, tone, expertise, clinical voice |
P2 | Specify Format | Tell the AI the exact SHAPE of the output — structure, length, fields |
P3 | Provide Examples | Show the AI WHAT good looks like — zero-shot, one-shot, few-shot |
P4 | Evaluate Quality | RATE and ITERATE — build rubrics, test variables, fix what fails |
P5 | Divide Labor | CHAIN prompts — output of Step 1 becomes input of Step 2 |
How the 5 Principles Layer Together
|