AI Prompting Cheat Sheet

A quick-reference guide to prompting patterns that work across ChatGPT, Claude, Gemini, and other AI models. Copy, paste, and customize.

AI Tutorials · · Updated · 2 min read

Quick answer

Effective AI prompts follow repeatable patterns: role assignment (tell the AI who to be), structured output (specify the format you want), few-shot examples (show what good output looks like), and chain-of-thought (ask it to reason step by step). This cheat sheet covers the most useful patterns across ChatGPT, Claude, and Gemini.

Quick Reference: Prompting Patterns

Role Assignment

You are a [role] with [years] of experience in [domain].
Your communication style is [style].

Structured Output

Provide your response in the following format:
## Summary
[2-3 sentence overview]
## Key Points
[Bulleted list]
## Recommendation
[Your top recommendation with reasoning]

Comparison Template

Compare [A] vs [B] for [use case].
Consider: [criteria 1], [criteria 2], [criteria 3].
Present as a table with ratings out of 5.

Code Generation

Write a [language] function that [description].
Requirements:
- [requirement 1]
- [requirement 2]
Include: type hints, error handling, docstring, and example usage.

Content Creation

Write a [content type] about [topic].
Target audience: [audience]
Tone: [tone]
Length: [word count]
Must include: [required elements]

Analysis Framework

Analyze [subject] using the following framework:
1. Current state assessment
2. Strengths and weaknesses
3. Opportunities and threats
4. Recommended actions (prioritized)

Iterative Refinement

Here's my draft: [paste draft]

Improve it by:
1. Making the language more concise
2. Adding specific examples
3. Strengthening the conclusion
Keep the same structure and tone.

Tips

  • Start broad, then narrow — Let the AI generate first, then refine
  • Use delimiters — Triple backticks, XML tags, or markdown headers to separate sections
  • Specify what NOT to do — “Don’t use jargon” is as useful as “use simple language”
  • Ask for alternatives — “Give me 3 different approaches to this problem”

Frequently asked questions

What are the best AI prompting techniques?
The most effective techniques are: role assignment (give the AI an expert persona), structured output (specify exact format), few-shot examples (show examples of what you want), chain-of-thought (ask it to think step by step), and iterative refinement (build on previous outputs).
Do prompting techniques work across all AI models?
Yes. The core patterns (role assignment, structured output, chain-of-thought) work across ChatGPT, Claude, Gemini, and other major models. Some models respond better to certain techniques, but the fundamentals are universal.
What is the most important prompting technique?
Being specific. The single biggest improvement most people can make is adding context: who you are, what format you want, what the output will be used for, and what good vs bad looks like. Specificity beats every other technique.

Want to keep learning?

Explore our guided learning paths or try building something with AI right now.

Enjoyed this article?

Subscribe for more AI insights delivered to your inbox every week.

No spam. Unsubscribe anytime.