You're reading from The GitHub Copilot Handbook A practical guide to transforming the software development life cycle with GitHub Copilot

Product type Paperback

Published in Nov 2025

Publisher Packt

ISBN-13 9781806116638

Length 290 pages

Edition 1st Edition

Tools

CoPilot

Concepts

Web Design

Authors (2):

Rob Bos

Randy Pagels

View More author details

Table of Contents (18) Chapters

Preface

1. What is GitHub Copilot?

2. GitHub Copilot Explained FREE CHAPTER

3. Getting Started with Generative AI

4. Choosing the Right GitHub Copilot Plan

5. Getting Started with GitHub Copilot

6. Mastering GitHub Copilot in Your IDE: Inline Suggestions, Chat, and Agent Mode

7. Going Beyond Code: Debugging, Terminal, and Collaboration with GitHub Copilot

8. Exploring GitHub Copilot Integrations

9. Collaborating with Copilot on GitHub.com: Issues, PRs, Reviews, and Coding Agent

10. Extending GitHub Copilot with the Model Context Protocol (MCP)

11. Getting the Most Out of GitHub Copilot

12. Navigating the GitHub Copilot Learning Curve

13. Building an Internal GitHub Copilot Community

14. Changing the Narrative: Reframing Engineering with AI

15. Unlock Your Exclusive Benefits

Unlock this Book’s Free Benefits in 3 Easy Steps

16. Other Books You May Enjoy

17. Index

Coding Agent on GitHub.com

The GitHub Copilot Coding Agent brings automation directly into the browser by letting you delegate entire issues to Copilot. Instead of drafting fixes or writing code yourself, you can assign a task in GitHub.com and Copilot will take it from there. The agent runs in a secure cloud environment, makes changes on a new branch, opens a draft PR, and tags you for review. This turns issues into actionable work items that can move forward without waiting for human availability, while still keeping you in control of the final review and merge.

Coding Agent versus Agent Mode

GitHub offers two related but distinct ways of handing off work to AI: the Coding Agent and Agent Mode. Both fall under the broader idea of agentic AI, where you describe a task and the system iterates until it produces a result. The difference lies in where the work happens and how it is applied: