Patterns in AI-Augmented Software Development

A pattern language for working with LLM-based coding assistants.

Introduction

About the pattern language

Get Early Access by Supporting Chiron Codex

Chiron Codex makes the pattern language for AI-augmented software development available for free,...

Accessing the Agent Skills and Sample Code Repositories

There are two companion repositories for this book. The Chiron Codex Agent Skills repository cont...

Prompt Management

Patterns you use to improve the quality of the prompts you send to the language model.

Extract Prompt

Intent When you need to correct the way an agent performs an intermediate task in a process, tur...

Record Prompt

Intent Get the model to write down what you should have told it, for it to do the job correctly ...

Context Management

Ensure the language model pays attention to information that's relevant to your task.

Clean Slate

Clean Slate Intent Clear out the model’s context before continuing a task, or starting a new ta...

Memory Management

Retain information for the coding assistant to use across multiple tasks.

Generate Documentation

Intent Prompt the model to describe the code it creates. Generate reading guides that help you u...

Task Management

Change the work you do, and the way you work, with your coding assistant.

You Reflect

You Reflect Intent Review the transcript of your interaction with an agent, recall the tense mo...

Defer the Details

Intent Don't try to fix everything in one sitting. Tell the model to record open questions so yo...

Discover Opportunities

Generate different solutions to solve your problem, and different problems to solve.

Ask for Alternatives

Intent Get the model to suggest multiple solutions to your problem. Compare benefits and drawbac...

Disclose Ambiguity

Intent Prompt the model to list the assumptions inherent in your prompt, or missing information ...

Quality Improvement

Ensuring a high standard in the output the model generates.

Call Out Error

Call Out Error Intent Tell the agent when it does something wrong, before you commit to the mis...

Re-Read Instructions

Re-Read Instructions Intent Tell the model to pay attention to its original prompt, and check i...