axiom-foundation-models

Implement on-device AI on Apple platforms using the Foundation Models framework.

Updated Dec 3, 2025
One-click install
npx skills add https://github.com/tuliopc23/flying-dutchman-app --skill axiom-foundation-models-tuliopc23
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: axiom-foundation-models
Source: https://github.com/tuliopc23/flying-dutchman-app/tree/main/.claude/skills/axiom-foundation-models
Command: npx skills add https://github.com/tuliopc23/flying-dutchman-app --skill axiom-foundation-models-tuliopc23

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the implementation of on-device AI using Apple's Foundation Models framework, preventing common issues like context overflow, UI blocking, incorrect model usage, and manual data parsing.

Core Features & Use Cases

  • On-Device AI Implementation: Seamlessly integrate AI features directly on Apple devices.
  • Context Overflow Prevention: Manages the model's context window to avoid errors.
  • UI Responsiveness: Ensures the user interface remains interactive during AI generation.
  • Structured Output: Facilitates the generation of structured data using @Generable.
  • Use Case: Automatically summarize lengthy user feedback reports directly on the user's iPhone without needing an internet connection, ensuring privacy and speed.

Quick Start

Check if Foundation Models are available on the current device before proceeding.

Frequently Asked Questions about axiom-foundation-models

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

FAQPage Schema
How do I implement on-device AI in SwiftUI without blocking the UI?

Implement on-device AI in SwiftUI without blocking the UI by leveraging Apple's Foundation Models framework, which ensures the user interface remains responsive during AI generation. It addresses UI blocking by managing generation tasks properly.

What is the best way to parse structured JSON output from Apple Foundation Models?

The best way to parse structured output from Apple Foundation Models is using the `@Generable` property wrapper, which facilitates structured data generation and eliminates the need for manual JSON parsing on iOS, macOS, and visionOS.

Can I use Apple Foundation Models for offline text summarization on iOS?

Yes, you can use Apple Foundation Models for offline text summarization on iOS. This on-device AI implementation allows you to process tasks like summarizing lengthy user feedback reports directly on the device without needing an internet connection.

How do I prevent context overflow when using on-device AI in iOS development?

Prevent context overflow when using on-device AI in iOS development by leveraging the Foundation Models framework, which manages the model's context window automatically to avoid errors and ensure stable generation during lengthy processing tasks.

Do I need an internet connection to use Apple Foundation Models for generable tasks?

No, you do not need an internet connection to use Apple Foundation Models for generable tasks. The framework enables on-device AI implementation, ensuring privacy and speed by processing structured output generation directly on the user's device.

Why should I check if Foundation Models are available before starting on-device AI implementation?

You should check if Foundation Models are available because on-device AI implementation requires specific hardware capabilities. Verifying availability ensures your iOS, macOS, or visionOS app gracefully handles devices that do not support the framework.