Introduction

When Apple introduced AI integration with Xcode 26, it greatly improved the experience and productivity for anyone building an app on any of Apple platforms. Xcode automatically added context for the AI agent to improve the quality of the responses, and it was able to directly modify the source code to implement new features or changes you’re asking for.

But with the introduction of the Agentic AI tools, it took this power in a whole new level. You can build complete apps, tell the agent to “fix the build errors” without the need to give it any additional context as it will fix the errors, rebuild the project for you and repeat until the build succeeds. All without your interference.

In this module you’ll:

• Learn how to set up the Agentic AI tools in Xcode’s settings pane.
• Walk through a starter project to identify where improvements can be made by an Agentic AI.
• Learn how Agentic AI is different from tools like ChatGPT, and how Model Context Providers help LLMs with a unified API to various tools.
• Learn how with just a few prompts to an Agentic AI, you can add new features to an existing project, and do so mostly hands-off.
• see how Xcode can read Xcode’s issues list, provide fixes until there are no more errors, and even add documentation to your project.