ax-foundation-models

ax-foundation-models is a skill for Claude Code from Kasempiternal/axiom-v2. It costs 39 tokens per session (2,656 once invoked), scanned A, original, MIT.

A guide to Apple’s Foundation Models framework, which runs AI on supported Apple devices and can return text, structured data, streamed results, or tool calls.

In plain words
What is it for?
Add local language-model features to iOS 26-and-later apps, such as text generation, structured records, streaming responses, and calls to app-defined tools.
Why use it?
It helps developers use on-device AI while handling sessions, context, typed results, and diagnostic details correctly.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the axiom plugin — 40 skills, 8 commands, 12 agents, 2 hooks shipped together

Good fit Add local language-model features to iOS 26-and-later apps, such as text generation, structured records, streaming responses, and calls to app-defined tools.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kasempiternal/axiom-v2/ax-foundation-models
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Kasempiternal/axiom-v2 --skill ax-foundation-models
Clone the repo
git clone --depth 1 https://github.com/Kasempiternal/axiom-v2

Made for: Claude Code.

Or install axiom, the plugin that ships this one along with the rest of its 40 skills, 8 commands, 12 agents, 2 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ax-foundation-models

README.md
[![agentmods](https://agentmods.dev/badge/skills/kasempiternal/axiom-v2/ax-foundation-models/github.svg)](https://agentmods.dev/skills/kasempiternal/axiom-v2/ax-foundation-models)
Your own site
<a href="https://agentmods.dev/skills/kasempiternal/axiom-v2/ax-foundation-models"><img src="https://agentmods.dev/badge/skills/kasempiternal/axiom-v2/ax-foundation-models/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ax-foundation-models

Your own site · 80×15
<a href="https://agentmods.dev/skills/kasempiternal/axiom-v2/ax-foundation-models"><img src="https://agentmods.dev/badge/skills/kasempiternal/axiom-v2/ax-foundation-models.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,656 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00039 $0.02656
Opus 5 $0.00019 $0.01328
Sonnet 5 $0.00008 $0.00531
Haiku 4.5 $0.00004 $0.00266

Measured 10d ago against content hash d4528d98adf0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

ax-foundation-models scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

axiom-plugin/skills/ax-foundation-models/SKILL.md · 300 lines

How it starts

The opening of the file, as written. The whole thing — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Foundation Models — On-Device AI for Apple Platforms

Quick Patterns

Basic Session

import FoundationModels

let session = LanguageModelSession(instructions: "You are a helpful assistant.")
let response = try await session.respond(to: userInput)
response.content // plain String

@Generable Structured Output

@Generable
struct Person {
    @Guide(description: "Full name")
    let name: String
    @Guide(.range(1...120))
    let age: Int
}

let response = try await session.respond(to: "Generate a person", generating: Person.self)
response.content // type-safe Person

Streaming

let stream = session.streamResponse(to: prompt, generating: Itinerary.self)
for try await partial in stream {
    self.itinerary = partial // PartiallyGenerated — all properties optional
}

Tool Calling

struct GetWeatherTool: Tool {
    let name = "getWeather"
    let description = "Retrieve latest weather for a city"

    @Generable
    struct Arguments {
        @Guide(description: "The city to fetch weather for")
        var city: String
    }

    func call(arguments: Arguments) async throws -> ToolOutput {
        let weather = try await WeatherService.shared.weather(for: /* geocoded location */)
        return ToolOutput("\(arguments.city): \(weather.currentWeather.temperature.value) degrees")
    }
}

let session = LanguageModelSession(tools: [GetWeatherTool()])

Availability Check (mandatory)

switch SystemLanguageModel.default.availability {
case .available:
    let session = LanguageModelSession()
case .unavailable(let reason):
    // Show: "AI features require Apple Intelligence"
}

Error Handling (mandatory)

do {
    let response = try await session.respond(to: prompt)
} catch LanguageModelSession.GenerationError.exceededContextWindowSize {
    session = condensedSession(from: session) // see Context Management
} catch LanguageModelSession.GenerationError.guardrailViolation {
    showMessage("I can't help with that request")
} catch LanguageModelSession.GenerationError.unsupportedLanguageOrLocale {
    showMessage("Language not supported")
}

Read the full file on GitHub · 300 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 300 lines · 39 tokens per session scan A d4528d98adf0

Subscribe to this mod's changes

ax-foundation-models is a skill published in the GitHub repository Kasempiternal/axiom-v2 (4 stars, last pushed 6mo ago), licensed MIT. It adds 39 tokens to every session and 2,656 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

coreml

Integrates and profiles Core ML models for on-device inference. Use for mlmodel/mlpackage loading, generated or feature-provider predictions, compute-unit selection, MLTensor, Vision integration, MLComputePlan, model pipelines, deployment, or performance analysis.

thiennc-tesoglobal/ios-skills · 53 tokens

apple-on-device-ai

Designs private on-device AI for Apple platforms with Foundation Models, Core ML, MLX Swift, or llama.cpp. Use for local LLM runtime selection, Apple Intelligence chat or tool use, Apple Silicon inference, model conversion/compression, or backend comparison; route Core ML prediction code to coreml.

thiennc-tesoglobal/ios-skills · 66 tokens

vision-framework

Builds or reviews iOS computer-vision features with Vision and VisionKit, including OCR, barcode and document scanning, face/object detection, segmentation, tracking, and Core ML inference. Use for Vision requests, live DataScanner flows, or Vision/Core ML integration.

thiennc-tesoglobal/ios-skills · 56 tokens

workflow-audit

Systematic UI workflow auditing for SwiftUI applications. Discovers entry points, traces user flows, detects dead ends and broken promises, audits data wiring, evaluates from user perspective. Triggers: "workflow audit", "audit flows", "find dead ends", "check navigation".

Terryc21/workflow-audit · 59 tokens

swift-expert

Expert-level Swift development for iOS, macOS with SwiftUI, Combine, and modern Swift 5.9+. Use when the user mentions iOS, macOS, SwiftUI, Combine, async await, or Apple platforms, or when the task involves Modern Swift Features, Basics and Optionals, Functions and Closures, or Structs and Classes.

personamanagmentlayer/pcl · 76 tokens

ios-expert

Expert in iOS development with SwiftUI, UIKit, Combine, and Apple ecosystem integration. Use when the user mentions mobile, Swift, SwiftUI, UIKit, Apple platforms, or Xcode, or when the task involves iOS App Architecture, SwiftUI Fundamentals, UIKit Essentials, or Combine Framework.

personamanagmentlayer/pcl · 64 tokens