4 Prompt Patterns
This chapter covers
- Eight foundational Prompt Patterns for common Language Model interaction problems
- Structural Patterns that establish consistent behavior and output quality
- Contextual Patterns that surface overlooked risks, gaps, and blind spots
- Transformational Patterns that improve content quality
- Combining Patterns to handle complex prompting problems
In Chapters 2 and 3, you mastered the building blocks of effective prompts: Structural Elements that organize information and Linguistic Elements that influence communication precision. These foundations enable you to construct well-formed prompts, but they don't address how to interact with Language Models strategically across different types of problems.
This chapter introduces Prompt Patterns, repeatable strategies for solving common prompting problems.
4.1 Understanding Prompt Patterns
Prompt Patterns are proven interaction methods for getting reliable outputs from Language Models. Like Design Patterns in Software Engineering, they provide reusable solutions to recurring problems. Unlike formatting choices, patterns define strategy—how the model should approach the task. They sit above Structural and Linguistic Elements and shape whether the model generates, critiques, or adapts. Their key advantage is transferability: once you learn a pattern like Self-Reflection, you can apply it across domains without changing the core method.
What You'll Learn
Eight patterns are organized into three categories: