Meta Prompting
Meta prompting is using an AI to write, improve, or evaluate a prompt rather than to do the underlying task directly. You prompt the model about prompting.
Why it matters to educators
Educators do not need to become prompt experts. The model already knows what a strong prompt looks like. Asking it to improve your prompt before you run the real task raises the quality of everything downstream: lesson materials, student feedback, family communication. It is the fastest prompting skill to pick up because the AI does the hard part.
The foundation
Meta prompting separates two jobs: writing the prompt and doing the task. Instead of asking for the lesson plan right away, you first ask the AI what it needs from you to build a good one, or you hand it your rough prompt and ask for a stronger version. Then you run the improved prompt.
It usually takes one of three forms:
"What information do you need from me to do this well?" "Rewrite my prompt to get a better result." "Here is my prompt and the output I got. What was missing?"
Meta prompting does not replace your judgment. You still decide whether the improved prompt reflects your instructional goal, and you still review the final output for accuracy and fit.
What this can look like in education
Build a reusable classroom prompt
A teacher drafts a rough prompt for generating exit tickets, then asks the AI to strengthen it and turn it into a template with blanks for topic and grade level. The refined prompt gets reused all year and shared with the department.