Implement Apple Intelligence features with LanguageModelSession and @Generable structured output.
npx skills add https://github.com/DengNaichen/Stet --skill axiom-ai
Please help me install this Agent Skill. Skill: axiom-ai Source: https://github.com/DengNaichen/Stet/tree/main/.agents/skills/axiom-ai Command: npx skills add https://github.com/DengNaichen/Stet --skill axiom-ai
This skill provides the definitive framework for implementing, testing, and evaluating Apple Intelligence and on-device AI features, preventing common architectural pitfalls and runtime failures.
Use the axiom-ai skill to implement a LanguageModelSession for structured data generation using the @Generable protocol.
High-intent search queries and answers about installing and using this skill.
To implement structured output with Apple Intelligence Foundation Models, use the @Generable protocol within a LanguageModelSession to define and generate typed data structures directly on-device.
The Tool protocol in on-device AI frameworks is an integration interface that allows the Foundation Model to invoke external functions and interact with app-specific logic during a LanguageModelSession.
To evaluate and calibrate LLM performance for on-device AI features, implement evaluation-driven development by designing datasets and calibrating model-as-judge evaluators to measure accuracy against expected outputs.
Yes, Apple Intelligence supports custom guardrails and context window management by requiring strict adherence to on-device AI framework constraints during LanguageModelSession implementation and model refinement.
@Generable model output fails during on-device inference when the LanguageModelSession does not properly enforce schema constraints or when context window limits truncate the Foundation Model's response generation.
RektooooooImplement on-device AI in iOS, macOS, and visionOS apps using Apple's Foundation Models framework.
Miosa-osaDesign binary pass/fail LLM-as-Judge evaluation prompts with scoring instructions and guardrails.
mjunaidcaConnect to MCP servers to discover and execute tools on demand.
rshankrasIntegrate on-device LLM text generation and tool calling via Apple's Foundation Models framework.
