foundation-models

Name: foundation-models
Author: rshankras

2installs46forks

How to install foundation-models?

rshankras/claude-code-apple-skills/foundation-models

$npx -y skills add rshankras/claude-code-apple-skills --skill foundation-models

Installs into the current project.

Summary

On-device LLM integration using Apple's Foundation Models framework. Use when implementing AI text generation, structured output, or tool calling.

Files · 1

View on GitHub

SKILL.md

1	# Foundation Models
2
3	Integrate Apple's on-device LLM into your apps for privacy-preserving AI features. Companion references: safety-and-guardrails.md (model limits, prompt design, the four-layer safety stack), models-and-agents.md (Private Cloud Compute, `LanguageModel` protocol, vision input, `DynamicProfile` agentic sessions — the iOS 27 wave), and utilities-package.md (Apple's open-source utilities package: OpenAI-compatible endpoints, just-in-time Skills, history compression).
4
5	## When This Skill Activates
6
7	- User wants AI text generation features
8	- User needs structured data from natural language
9	- User asks about prompting or LLM integration
10	- User wants to implement AI assistants or agentic features (tool loops, multi-profile sessions)
11	- User needs content summarization or extraction
12	- User asks about Private Cloud Compute, guardrails, or model safety
13
14	## Model Fit — Check Before Building
15
16	The on-device model is ~3B parameters (2-bit quantized): built for summarization, extraction, classification, tagging, revision, short chat — not math, code generation, facts, or world knowledge (WWDC25 248). For capability boundaries, prompt-design rules, and the safety stack, read `safety-and-guardrails.md` first. For anything bigger, `PrivateCloudComputeLanguageModel` (32k context, reasoning) and third-party backends are in `models-and-agents.md`.
17
18	## Quick Start
19
20	### 1. Check Availability
21
22	```swift
23	import FoundationModels
24
25	struct IntelligentView: View {
26	private var model = SystemLanguageModel.default
27
28	var body: some View {
29	switch model.availability {
30	case .available:
31	ContentView()
32	case .unavailable(.deviceNotEligible):
33	UnsupportedDeviceView()
34	case .unavailable(.appleIntelligenceNotEnabled):
35	EnableIntelligenceView()
36	case .unavailable(.modelNotReady):
37	ModelDownloadingView()
38	case .unavailable(let reason):
39	ErrorView(reason: reason)
40	}
41	}
42	}
43	```
44
45	### 2. Create a Session
46
47	```swift
48	// Simple session
49	let session = LanguageModelSession()
50
51	// Session with instructions
52	let session = LanguageModelSession(instructions: """
53	You are a helpful cooking assistant.
54	Provide concise, practical advice for home cooks.
55	""")
56	```
57
58	### 3. Generate Response
59
60	```swift
61	let response = try await session.respond(to: "What's a quick dinner idea?")
62	print(response.content)
63	```
64
65	## Prompt Engineering Best Practices
66
67	### The Instruction Formula
68
69	Instructions set the model's persona and constraints. They're prioritized over prompts.
70
71	```
72	[Role] + [Task] + [Style] + [Safety]
73	```
74
75	Example:
76	```swift
77	let instructions = """
78	You are a fitness coach specializing in home workouts.
79	Help users create exercise routines based on their equipment and goals.
80	Keep responses under 100 words and use bullet points for exercises.
81	Decline requests for medical advice and suggest consulting a doctor.
82	"""
83	```
84
85	### Instruction Components
86
87	\| Component \| Purpose \| Example \|
88	\|-----------\|---------\|---------\|
89	\| Role \| Define persona \| "You are a travel expert" \|
90	\| Task \| What to do \| "Help plan itineraries" \|
91	\| Style \| Output format \| "Use bullet points, be concise" \|
92	\| Safety \| Boundaries \| "Don't provide medical advice" \|
93
94	### Effective Prompts
95
96	Prompts are user inputs. Make them:
97
98	\| Principle \| Bad \| Good \|
99	\|-----------\|-----\|------\|
100	\| Specific \| "Help with cooking" \| "Suggest a 30-minute vegetarian dinner" \|
101	\| Constrained \| "Tell me about dogs" \| "Describe Golden Retrievers in 3 sentences" \|
102	\| Focused \| "I need help with many things" \| "What ingredients substitute for eggs in baking?" \|
103
104	### Prompt Patterns
105
106	Question Pattern:
107	```swift
108	let prompt = "What are three ways to reduce food waste at home?"
109	```
110
111	Command Pattern:
112	```swift
113	let prompt = "Create a weekly meal plan for a family of four, budget-friendly."
114	```
115
116	Extraction Pattern:
117	```swift
118	let prompt = """
119	Extract the following from this email:
120	- Sender name
121	- Meeting date
122	- Action items
123
124	Email: \(emailContent)
125	"""
126	```
127
128	Transformation Pattern:
129	```swift
130	let prompt = "Rewrite this text to be more formal: \(casualText)"
131	```
132
133	## Structured Output with @Generable
134
135	Get typed Swift data instead of raw strings.
136
137	### Define Generable Types
138
139	```swift
140	@Generable(description: "A recipe suggestion")
141	struct Recipe {
142	var name: String
143
144	@Guide(description: "Cooking time in minutes", .range(5...180))
145	var cookingTime: Int
146
147	@Guide(description: "Difficulty level", .options(["Easy", "Medium", "Hard"]))
148	var difficulty: String
149
150	@Guide(description:

People who liked this also viewed

6 picks

Type

hyperframes-cliUse the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play, present, beats, keyframes, single or batch render, publish, cloud, SkillsJul 2026262k37k
request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026165k181k
analyze-projectRigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrSkillsMay 2026161k507
explore-codeRigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated braSkillsMay 2026161k507
ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026160k181k
scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026149k181k

foundation-models

2installs46forks

View on GitHub

How to install foundation-models?

rshankras/claude-code-apple-skills/foundation-models

$npx -y skills add rshankras/claude-code-apple-skills --skill foundation-models

Installs into the current project.

Summary

On-device LLM integration using Apple's Foundation Models framework. Use when implementing AI text generation, structured output, or tool calling.

Files · 1

View on GitHub

SKILL.md

1	# Foundation Models
2
3	Integrate Apple's on-device LLM into your apps for privacy-preserving AI features. Companion references: safety-and-guardrails.md (model limits, prompt design, the four-layer safety stack), models-and-agents.md (Private Cloud Compute, `LanguageModel` protocol, vision input, `DynamicProfile` agentic sessions — the iOS 27 wave), and utilities-package.md (Apple's open-source utilities package: OpenAI-compatible endpoints, just-in-time Skills, history compression).
4
5	## When This Skill Activates
6
7	- User wants AI text generation features
8	- User needs structured data from natural language
9	- User asks about prompting or LLM integration
10	- User wants to implement AI assistants or agentic features (tool loops, multi-profile sessions)
11	- User needs content summarization or extraction
12	- User asks about Private Cloud Compute, guardrails, or model safety
13
14	## Model Fit — Check Before Building
15
16	The on-device model is ~3B parameters (2-bit quantized): built for summarization, extraction, classification, tagging, revision, short chat — not math, code generation, facts, or world knowledge (WWDC25 248). For capability boundaries, prompt-design rules, and the safety stack, read `safety-and-guardrails.md` first. For anything bigger, `PrivateCloudComputeLanguageModel` (32k context, reasoning) and third-party backends are in `models-and-agents.md`.
17
18	## Quick Start
19
20	### 1. Check Availability
21
22	```swift
23	import FoundationModels
24
25	struct IntelligentView: View {
26	private var model = SystemLanguageModel.default
27
28	var body: some View {
29	switch model.availability {
30	case .available:
31	ContentView()
32	case .unavailable(.deviceNotEligible):
33	UnsupportedDeviceView()
34	case .unavailable(.appleIntelligenceNotEnabled):
35	EnableIntelligenceView()
36	case .unavailable(.modelNotReady):
37	ModelDownloadingView()
38	case .unavailable(let reason):
39	ErrorView(reason: reason)
40	}
41	}
42	}
43	```
44
45	### 2. Create a Session
46
47	```swift
48	// Simple session
49	let session = LanguageModelSession()
50
51	// Session with instructions
52	let session = LanguageModelSession(instructions: """
53	You are a helpful cooking assistant.
54	Provide concise, practical advice for home cooks.
55	""")
56	```
57
58	### 3. Generate Response
59
60	```swift
61	let response = try await session.respond(to: "What's a quick dinner idea?")
62	print(response.content)
63	```
64
65	## Prompt Engineering Best Practices
66
67	### The Instruction Formula
68
69	Instructions set the model's persona and constraints. They're prioritized over prompts.
70
71	```
72	[Role] + [Task] + [Style] + [Safety]
73	```
74
75	Example:
76	```swift
77	let instructions = """
78	You are a fitness coach specializing in home workouts.
79	Help users create exercise routines based on their equipment and goals.
80	Keep responses under 100 words and use bullet points for exercises.
81	Decline requests for medical advice and suggest consulting a doctor.
82	"""
83	```
84
85	### Instruction Components
86
87	\| Component \| Purpose \| Example \|
88	\|-----------\|---------\|---------\|
89	\| Role \| Define persona \| "You are a travel expert" \|
90	\| Task \| What to do \| "Help plan itineraries" \|
91	\| Style \| Output format \| "Use bullet points, be concise" \|
92	\| Safety \| Boundaries \| "Don't provide medical advice" \|
93
94	### Effective Prompts
95
96	Prompts are user inputs. Make them:
97
98	\| Principle \| Bad \| Good \|
99	\|-----------\|-----\|------\|
100	\| Specific \| "Help with cooking" \| "Suggest a 30-minute vegetarian dinner" \|
101	\| Constrained \| "Tell me about dogs" \| "Describe Golden Retrievers in 3 sentences" \|
102	\| Focused \| "I need help with many things" \| "What ingredients substitute for eggs in baking?" \|
103
104	### Prompt Patterns
105
106	Question Pattern:
107	```swift
108	let prompt = "What are three ways to reduce food waste at home?"
109	```
110
111	Command Pattern:
112	```swift
113	let prompt = "Create a weekly meal plan for a family of four, budget-friendly."
114	```
115
116	Extraction Pattern:
117	```swift
118	let prompt = """
119	Extract the following from this email:
120	- Sender name
121	- Meeting date
122	- Action items
123
124	Email: \(emailContent)
125	"""
126	```
127
128	Transformation Pattern:
129	```swift
130	let prompt = "Rewrite this text to be more formal: \(casualText)"
131	```
132
133	## Structured Output with @Generable
134
135	Get typed Swift data instead of raw strings.
136
137	### Define Generable Types
138
139	```swift
140	@Generable(description: "A recipe suggestion")
141	struct Recipe {
142	var name: String
143
144	@Guide(description: "Cooking time in minutes", .range(5...180))
145	var cookingTime: Int
146
147	@Guide(description: "Difficulty level", .options(["Easy", "Medium", "Hard"]))
148	var difficulty: String
149
150	@Guide(description:

People who liked this also viewed

6 picks

Type