macos-tahoe-apis

Guides implementation of macOS 26 Tahoe APIs including Apple Intelligence, MLX, and Continuity.

70|4|Updated Jul 22, 2025
One-click install
npx skills add https://github.com/AlexW00/Zettel --skill macos-tahoe-apis-alexw00
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: macos-tahoe-apis
Source: https://github.com/AlexW00/Zettel/tree/main/.agents/skills/macos-development/macos-tahoe-apis
Command: npx skills add https://github.com/AlexW00/Zettel --skill macos-tahoe-apis-alexw00

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Developers targeting macOS 26 (Tahoe) often struggle to adopt new platform APIs correctly, missing modern features like Apple Intelligence, MLX machine learning, and Continuity integration while relying on deprecated approaches. ## Core Features & Use Cases - Tahoe Feature Guidance: Covers redesigned Spotlight integration, Phone app links, Control Center widgets, custom folder icons, and urgent reminders with Swift code examples. - Apple Intelligence & MLX: Provides patterns for Foundation Models, on-device AI inference, MCP tool integration, and MLX training with M5 neural accelerator support. - Continuity Integration: Explains Universal Clipboard, Handoff via NSUserActivity, and AirDrop sharing between macOS, iOS, and iPadOS. - Use Case: When adding Spotlight search indexing to a Mac note-taking app, consult this Skill to get the correct CoreSpotlight CSSearchableItem implementation and quick action registration code. ## Quick Start Ask how to implement a specific macOS 26 feature such as Spotlight indexing or Handoff in your Swift app.

Frequently Asked Questions about macos-tahoe-apis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I integrate Spotlight search in a macOS app?▼

Use CoreSpotlight to index content with CSSearchableItem and CSSearchableItemAttributeSet containing title, description, and keywords. Register quick actions via NSUserActivity with isEligibleForSearch enabled, then handle selections in the application continue userActivity delegate method.

How to use Apple Intelligence Foundation Models in Swift?▼

Load a Foundation Model for tasks like text generation or summarization, then call its generate or summarize methods asynchronously. Set processingLocation to onDevice to ensure privacy-preserving inference without sending data off the device.

Does MLX support the M5 chip neural engine?▼

Yes, MLX can leverage M5 neural accelerators by setting MLX.configuration.useNeuralAccelerators to true and preferredDevice to neuralEngine. You can check availability with MLX.neuralAcceleratorsAvailable before running inference or training.

How do I implement Handoff between macOS and iOS?▼

Create an NSUserActivity with a custom activityType, set isEligibleForHandoff to true, and call becomeCurrent. On the receiving device, implement the application continue userActivity delegate method to restore state from the userInfo dictionary.

What are the limitations of Control Center widgets on macOS?▼

Control Center customization is system-level, so apps can only provide widgets through WidgetKit using StaticConfiguration with supported families like systemSmall. Apps cannot directly add custom controls to Control Center outside the WidgetKit mechanism.