Skip to content

This repository is a place for experimenting with LLM prompts using prompt engineering patterns. The content of this repo is heavily based on Jules' White course "Prompt Engineering for ChatGPT"

License

Notifications You must be signed in to change notification settings

trinchetto/prompt-engineering-patterns-handbook

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Prompt Engineering Patterns Handbook

Introduction

Prompt engineering is the art of crafting clear, purposeful instructions for large language models (LLMs) like ChatGPT. Done well, it’s less “asking nicely” and more architecting conversations so the AI delivers exactly what you need — consistently.

The patterns here are reusable blueprints. Each is like a Lego brick: snap them together to build anything from a polite email rewriter to a fully automated AI agent workflow.

Overview

The Prompt Engineering Patterns Handbook is a Markdown-based guide that catalogues reusable prompt-design blueprints for large language models. The repository includes a project overview README.md and a docs/ directory with four themed pattern files (docs/interaction-patterns.md, docs/structuring-patterns.md, docs/reasoning-patterns.md, and docs/creativity-patterns.md) alongside contributing guidelines.

Architecture & Concepts

  • Master indexdocs/README.md lists every pattern, grouped into four categories: Interaction, Structuring, Reasoning & Analysis, and Creativity & Exploration.
  • Pattern files – Each category file in docs/ documents specific patterns using a consistent template: description, purpose, problem solved, data needed, a “quick copy-paste prompt,” and general vs. AI-agent examples.
  • Custom GPTsgpts/ contains ready-to-use GPT configurations that apply these patterns in practice.
  • Contribution workflowCONTRIBUTING.md outlines how to add a new pattern: choose the correct group file, follow the standard section format, and update the docs catalog.

Pointers for Further Exploration

  1. Dive into pattern groups: Start with the docs’ “Detailed Guides” links to explore each category in depth.
  2. Practice with examples: Try out the “Quick Copy-Paste Prompt” snippets in the pattern files to see how they influence model responses.
  3. Contribute new patterns: Follow the contributing guidelines to add or refine patterns, keeping examples and prompts concise.
  4. Experiment with chaining patterns: Combine multiple patterns (e.g., Persona + Template) to build richer, more controllable prompt workflows.

Explore the full pattern catalog and detailed guides in the docs directory.

About

This repository is a place for experimenting with LLM prompts using prompt engineering patterns. The content of this repo is heavily based on Jules' White course "Prompt Engineering for ChatGPT"

Resources

License

Contributing

Stars

Watchers

Forks

Packages

No packages published

Languages