apple-intelligence
installsforks
How to install apple-intelligence?
rshankras/claude-code-apple-skills/apple-intelligence
$npx -y skills add rshankras/claude-code-apple-skills --skill apple-intelligenceApple Intelligence skills for on-device AI features including Foundation Models, Visual Intelligence, App Intents, and intelligent assistants. Use when implementing AI-powered features.
SKILL.md
| 1 | # Apple Intelligence Skills |
| 2 | |
| 3 | Skills for implementing Apple Intelligence features including on-device LLMs, visual recognition, App Intents integration, and intelligent assistants. |
| 4 | |
| 5 | ## When This Skill Activates |
| 6 | |
| 7 | Use this skill when the user: |
| 8 | - Wants to add AI/LLM features to their app |
| 9 | - Needs on-device text generation or understanding |
| 10 | - Asks about Foundation Models or Apple Intelligence |
| 11 | - Wants to implement structured AI output |
| 12 | - Needs prompt engineering guidance |
| 13 | - Wants camera-based visual intelligence features |
| 14 | - Needs Siri, Shortcuts, or Spotlight integration via App Intents |
| 15 | - Wants to expose app actions or content to the system |
| 16 | |
| 17 | ## Available Skills |
| 18 | |
| 19 | ### foundation-models/ |
| 20 | On-device LLM integration with prompt engineering best practices. |
| 21 | - Model availability checking |
| 22 | - Session management |
| 23 | - @Generable structured output (property-order + schema-injection rules) |
| 24 | - Tool calling patterns |
| 25 | - Snapshot streaming |
| 26 | - Prompt engineering techniques |
| 27 | - `safety-and-guardrails.md` — model limits, the instructions-over-prompts hierarchy, guardrail handling, the four-layer safety stack, evals |
| 28 | - `models-and-agents.md` — Private Cloud Compute, LanguageModel protocol (third-party backends), vision input, DynamicProfile agents, tool-calling modes, KV-cache rules |
| 29 | |
| 30 | ### visual-intelligence/ |
| 31 | Integrate with iOS Visual Intelligence for camera-based search. |
| 32 | - IntentValueQuery implementation |
| 33 | - SemanticContentDescriptor handling |
| 34 | - AppEntity for searchable content |
| 35 | - Display representations |
| 36 | - Deep linking from results |
| 37 | |
| 38 | ### app-intents/ |
| 39 | App Intents for Siri, Shortcuts, Spotlight, and Apple Intelligence. |
| 40 | - AppIntent protocol, parameters, perform() |
| 41 | - AppEntity and entity queries |
| 42 | - App Shortcuts with voice phrases |
| 43 | - IndexedEntity and Spotlight indexing |
| 44 | - Intent modes (background, foreground) |
| 45 | - Interactive snippets with SnippetIntent |
| 46 | - Visual intelligence integration |
| 47 | - Onscreen entities for Siri/ChatGPT |
| 48 | - Multiple choice API |
| 49 | - Swift package support |
| 50 | |
| 51 | ## Key Principles |
| 52 | |
| 53 | ### 1. Privacy First |
| 54 | - On-device processing by default; Private Cloud Compute extends capacity without storing prompts (no account/API keys — WWDC26 241) |
| 55 | - When routing to less-private backends, redact context first (see foundation-models/models-and-agents.md) |
| 56 | |
| 57 | ### 2. Graceful Degradation |
| 58 | - Always check model availability |
| 59 | - Provide fallback UI for unsupported devices |
| 60 | - Handle errors gracefully (incl. guardrail violations — silent for proactive features, explained for user-initiated) |
| 61 | |
| 62 | ### 3. Efficient Prompting |
| 63 | - Keep prompts focused and specific |
| 64 | - Use structured output when possible |
| 65 | - Respect context window limits — query `model.contextSize`, never hardcode (4,096 first-gen on-device; 8,192 iOS 27 on-device; 32,000 Private Cloud Compute) |
| 66 | |
| 67 | ## Reference Documentation |
| 68 | |
| 69 | - [Foundation Models](https://developer.apple.com/documentation/foundationmodels) |
| 70 | - [Visual Intelligence](https://developer.apple.com/documentation/visualintelligence) |
| 71 | - [App Intents](https://developer.apple.com/documentation/appintents) |
| 72 | - Local captured docs (optional): if `~/Downloads/docs/` contains `FoundationModels-Using-on-device-LLM-in-your-app.md`, `Implementing-Visual-Intelligence-in-iOS.md`, or `AppIntents-Updates.md`, read them for extra detail; skip silently if absent. |
People who liked this also viewed
6 picksType
- 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 2026
- request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
- 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 2026
- 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 2026
- ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
- scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
