Claude Code: Up and Running

Chapter 3. Talking to Claude Code

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 3rd chapter of the final book. The GitHub repo is active.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

Chapter 2 got Claude Code running. This chapter is about getting better work out of it.

Claude Code is different from the basic Claude desktop chat application or the web site, because it’s not just a search box that hands back an answer. It behaves more like a teammate¹ sitting inside your project, able to read your files, execute commands, and change the code in the project.

This chapter is about using prompts to direct the process. Some prompts are questions, for when you are still figuring out what to do. Some are requests, for when you’re ready to implement. Along the way we’ll look at how to pick a conversation back up later, and how to keep Claude Code from guessing when judgment is required.

Prompts as Work Requests

Short, conversational prompts have their place. Chapter 2 suggested some options when you first encounter an existing project you haven’t used, like “What does this project do?” or “Where is the main entry point?” or “What technology stack does this system use?” That is ...