With Xcode version 26.3, Apple is adding support for coding agents, allowing developers to use tools like Claude Agent and OpenAI's Codex directly in Xcode to build apps.
See also: Fake Moltbot AI coding assistant on VS Code Marketplace
Coding with agents will allow Xcode to complete more complex app development tasks autonomously. AI models like Claude, ChatGPT, and others have been available for use in Xcode since Apple added intelligence features in Xcode 26, but until now the AI has been limited and unable to take action on its own. That will change with the option to use an AI coding assistant.
AI models can access more of Xcode’s features to help them achieve a project goal, and Apple worked directly with Anthropic and OpenAI to configure their agents for use in Xcode. Agents can create new files, examine the structure of a project in Xcode, build a project directly and run tests, take screenshots to check their work, and access Apple’s comprehensive developer documentation designed for AI agents.
Adding an agent to Xcode can be done with a single click in Xcode settings, with agents being able to update automatically as AI companies release updates. Developers will need to create an account with Anthropic or OpenAI to use these coding tools in Xcode, paying fees based on API usage.
See also: Moltbot, the AI agent that "really does things"
Apple says its goal was to ensure that Claude Agent and Codex operate efficiently, with reduced token usage. It's simple to switch between agents in the same project, giving developers the flexibility to choose the agent best suited for a particular task.
While Apple has partnered with OpenAI and Anthropic to integrate into Xcode, the features in Xcode 26.3 can be used with any agent or tool that uses the open Model Context Protocol standard. Apple is releasing documentation so developers can configure and connect MCP agents in Xcode.
Using natural language commands, developers can instruct AI agents to complete a task, such as adding a new feature to an app. Xcode then works with the agent to break down the instructions into small tasks, and the agent can work on its own from there.
In a project's sidebar, developers can track what the agent is doing using the transcript, and they can click to see where code is added to track what the agent is doing. At any point, developers can go back before an agent or model makes a change, so there are options to undo unwanted results or try multiple options for introducing a new feature.
See also: Google Chrome: New 'auto browse' feature via Gemini AI
Apple says that coding with agents will allow developers to simplify workflows, make changes faster, and bring new ideas to life. Apple also sees it as a learning tool that gives developers the opportunity to learn new ways to build something or implement an API in an app.
