Meta-Prompting: Prompts That Write Prompts
The fastest way to a great prompt is asking the model to write and critique one.
The prompt
You are a prompt engineer. Task: I need a reusable prompt that turns raw meeting notes into a client-ready status email. Draft the prompt, then critique it for 3 likely failure modes (missing info, wrong tone, hallucinated commitments), then output the improved final prompt in a code block.
Pro tip: Always ask for the critique step — the first draft prompt is rarely the best one, and the model knows its own failure modes.
How to use this prompt
- Copy the prompt using the button above.
- Add any context specific to your situation — the more specific you are, the better the output.
- Paste it into Claude and run it.
- If the answer feels generic, add constraints: audience, length, tone, and what to avoid. That single change fixes most weak output.
Learn the technique
Meta-Prompting: Prompts That Write Prompts
Module 5 — Expert: Prompt Systems · Prompt Engineering Fundamentals
Related prompts
The 5-Part Prompt Framework — ChatGPT
Every strong prompt combines five parts: Role (who the AI should act as), Context (background info it needs), Task (the specific ask), Format (how you want the output structured), and Constraints (length, tone, things to avoid).
ClaudeThe 5-Part Prompt Framework — Claude
Every strong prompt combines five parts: Role (who the AI should act as), Context (background info it needs), Task (the specific ask), Format (how you want the output structured), and Constraints (length, tone, things to avoid).
ChatGPTZero-Shot vs Few-Shot Prompting — ChatGPT
Zero-shot means asking directly with no examples.
GrokZero-Shot vs Few-Shot Prompting — Grok
Zero-shot means asking directly with no examples.
ChatGPTChain-of-Thought Prompting — ChatGPT
Asking a model to "think step by step" before answering measurably improves accuracy on reasoning, math, and multi-step logic tasks because it forces the model to externalize intermediate steps instead of jumping to a guess..
ClaudeChain-of-Thought Prompting — Claude
Asking a model to "think step by step" before answering measurably improves accuracy on reasoning, math, and multi-step logic tasks because it forces the model to externalize intermediate steps instead of jumping to a guess..
Topics
Related across the site
Reasoning model
A reasoning model is trained to generate extended internal deliberation before answering. It trades latency and cost for substantially better performance on maths, logic, coding and multi-step problems.
Zero-shot prompting
Zero-shot prompting means asking a model to do a task with no worked examples — just an instruction. It works well for common tasks the model has seen extensively in training.
Meta-prompting
Meta-prompting is using a model to write, critique or improve prompts. You describe the task and the failure modes you want to avoid, and the model drafts the prompt.
Build a Meeting Notes System
Turn rambling notes into decisions, owners and deadlines — in a format consistent enough that you can search it six months later. The value is in the structure, not the summary.
Build a Personal Study Assistant
Turn a stack of lecture notes into a study partner that quizzes you, explains what you got wrong, and tracks which topics you keep failing. You will finish with a reusable system prompt you can point at any subject.
Build Your Own Prompt Library
Stop rewriting the same prompt. Build a small, organised, tested personal library — and learn the versioning habit that stops it rotting into a folder of near-duplicates.