Is my source code encrypted before it leaves my Mac?

Yes. The Privatemode proxy runs locally on your Mac and encrypts all code context before it leaves your machine. It also performs remote attestation to verify the integrity of the Privatemode service, ensuring end-to-end protection for every AI request Xcode makes.

Does Privatemode store or learn from my code?

No. Privatemode never learns from your data. All requests are processed using confidential computing, and no code, prompts, or responses are retained after processing. This is enforced by hardware, not just policy.

How does this compare to Xcode’s default ChatGPT integration?

Xcode’s default Intelligence provider is ChatGPT, so OpenAI’s models process your code on OpenAI’s servers. With Privatemode, open-weight models do that work instead, and your code is encrypted before it leaves your Mac and processed inside a hardware-isolated environment whose memory stays encrypted, so by design neither the infrastructure provider nor Privatemode can read it.

Do I need to install any Xcode extensions or plugins?

No. Xcode has built-in support for custom OpenAI-compatible model providers. You simply add Privatemode as a provider in Settings > Intelligence in Xcode 26 or later. No extensions, plugins, or workarounds are required.

Integrating Privatemode AI with Xcode

Xcode's Intelligence features are wired to an external provider out of the box. Privatemode is a provider you add in settings, so the same assistance runs on open-weight models instead of OpenAI's or Anthropic's models, in an environment whose memory stays encrypted while it works on your code.

Introduction

Xcode Intelligence on a model you choose

The challenge

Intelligence sends your code to an external provider

Swift Assist, code generation, and agentic changes use an external provider unless you change it. For Apple platform work that code often concerns unreleased hardware, seeded OS builds, and client projects under NDA, and a conventional provider processes it in plaintext on its servers.

The solution

A provider you add, not a plugin you install

Xcode's model provider list is part of the IDE, so once the Privatemode proxy runs on your Mac, adding it takes a few fields in Settings. The proxy holds your key, verifies the service through remote attestation, and encrypts each request before it leaves the machine.

Benefits

Why use Privatemode AI with Xcode?

Built in, nothing to install

End-to-end confidentiality

Learn how

Free tier available

How to get started

How to set up Privatemode in Xcode

Get your API key

If you don’t have a Privatemode API key yet, you can generate one for free here.

Run the Privatemode proxy

The proxy verifies the integrity of the Privatemode service using confidential computing-based remote attestation. The proxy also encrypts all data before sending and decrypts data it receives.

Add Privatemode as a model provider in Xcode

Open Xcode 26 or later and go to Settings > Intelligence. Click Add a Model Provider, choose Locally Hosted, and enter localhost:8080 with a description such as Privatemode. Xcode adds the /v1 path itself, and no API key is needed here because the proxy holds your real Privatemode key.

Walk through the interactive demo

FAQ

Frequently asked questions about using Privatemode with Xcode

All of Xcode’s Intelligence features work with Privatemode, including Swift Assist, code generation, and agentic coding. Since Xcode 26 natively supports custom OpenAI-compatible model providers via Settings > Intelligence, every AI feature routes through Privatemode automatically once configured.