做什么
苹果基金会 iOS 26+中的device LLM的模型框架——文本生成,有指引的生成,使用@Genebable,工具调用和快照流
技能库 智客分类:Agent 工作流 foundation-models-on-device
苹果基金会 用于device LLM上的模型框架——文本生成,用@Genebable来引导生成,工具调用,以及快照流于iOS 26+. 在iOS 26+上与 Apple Foundation Models 添加上-device LLM 特性时使用.
官方网址:skills.sh
先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。
苹果基金会 iOS 26+中的device LLM的模型框架——文本生成,有指引的生成,使用@Genebable,工具调用和快照流
- 用于对隐私敏感的应用程序的在线设备文本生成
按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:FoundationModels: On-Device LLM (iOS 26)、When to Activate、Core Pattern — Availability Check、Core Pattern — Basic Session、Core Pattern — Guided Generation with @Generable、1. Define a Generable Type。
文件分析:这是一份仅含 SKILL.md 的指令型技能,代理激活后整份正文进入上下文。
Apple FoundationModels framework for on-device LLM — text generation, guided generation with @Generable, tool calling, and snapshot streaming in iOS 26+. Use when adding on-device LLM features with Apple FoundationModels on iOS 26+.
FoundationModels: On-Device LLM (iOS 26)When to ActivateCore Pattern — Availability CheckCore Pattern — Basic SessionCore Pattern — Guided Generation with @Generable1. Define a Generable Type2. Request Structured OutputSupported @Guide ConstraintsCore Pattern — Tool Calling1. Define a Tool2. Create Session with Tools3. Handle Tool Errors
来源分类:skills.sh agent-skill
namefoundation-models-on-devicedescription具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗
先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。
复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。
把 Agent Skill「foundation-models-on-device」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-b8c4b66e3e3b9648-%E8%AE%BE%E5%A4%87%E4%B8%8A%E7%9A%84%E5%9F%BA%E7%A1%80%E6%A8%A1%E5%9E%8B.html 请存为 .cursor/skills/foundation-models-on-device/SKILL.md 或 .claude/skills/foundation-models-on-device/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。
需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。
npx skills add 'https://github.com/affaan-m/ecc' --list
npx skills add 'https://github.com/affaan-m/ecc' --skill 'foundation-models-on-device'
CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。
Patterns for integrating Apple's on-device language model into apps using the FoundationModels framework. Covers text generation, structured output with @Generable, custom tool calling, and snapshot streaming — all running on-device for privacy and offline support.
Always check model availability before creating a session:
struct GenerativeView: View {
private var model = SystemLanguageModel.default
var body: some View {
switch model.availability {
case .available:
ContentView()
case .unavailable(.deviceNotEligible):
Text("Device not eligible for Apple Intelligence")
case .unavailable(.appleIntelligenceNotEnabled):
Text("Please enable Apple Intelligence in Settings")
case .unavailable(.modelNotReady):
Text("Model is downloading or not ready")
case .unavailable(let other):
Text("Model unavailable: \(other)")
}
}
}
// Single-turn: create a new session each time
let session = LanguageModelSession()
let response = try await session.respond(to: "What's a good month to visit Paris?")
print(response.content)
// Multi-turn: reuse session for conversation context
let session = LanguageModelSession(instructions: """
You are a cooking assistant.
Provide recipe suggestions based on ingredients.
Keep suggestions brief and practical.
""")
let first = try await session.respond(to: "I have chicken and rice")
let followUp = try await session.respond(to: "What about a vegetarian option?")
Key points for instructions:
Generate structured Swift types instead of raw strings:
@Generable(description: "Basic profile information about a cat")
struct CatProfile {
var name: String
@Guide(description: "The age of the cat", .range(0...20))
var age: Int
@Guide(description: "A one sentence profile about the cat's personality")
var profile: String
}
let response = try await session.respond(
to: "Generate a cute rescue cat",
generating: CatProfile.self
)
// Access structured fields directly
print("Name: \(response.content.name)")
print("Age: \(response.content.age)")
print("Profile: \(response.content.profile)")
.range(0...20) — numeric range.count(3) — array element countdescription: — semantic guidance for generationLet the model invoke custom code for domain-specific tasks:
struct RecipeSearchTool: Tool {
let name = "recipe_search"
let description = "Search for recipes matching a given term and return a list of results."
@Generable
struct Arguments {
var searchTerm: String
var numberOfResults: Int
}
func call(arguments: Arguments) async throws -> ToolOutput {
let recipes = await searchRecipes(
term: arguments.searchTerm,
limit: arguments.numberOfResults
)
return .string(recipes.map { "- \($0.name): \($0.description)" }.joined(separator: "\n"))
}
}
let session = LanguageModelSession(tools: [RecipeSearchTool()])
let response = try await session.respond(to: "Find me some pasta recipes")
do {
let answer = try await session.respond(to: "Find a recipe for tomato soup.")
} catch let error as LanguageModelSession.ToolCallError {
print(error.tool.name)
if case .databaseIsEmpty = error.underlyingError as? RecipeSearchToolError {
// Handle specific tool error
}
}
Stream structured responses for real-time UI with PartiallyGenerated types:
@Generable
struct TripIdeas {
@Guide(description: "Ideas for upcoming trips")
var ideas: [String]
}
let stream = session.streamResponse(
to: "What are some exciting trip ideas?",
generating: TripIdeas.self
)
for try await partial in stream {
// partial: TripIdeas.PartiallyGenerated (all properties Optional)
print(partial)
}
@State private var partialResult: TripIdeas.PartiallyGenerated?
@State private var errorMessage: String?
var body: some View {
List {
ForEach(partialResult?.ideas ?? [], id: \.self) { idea in
Text(idea)
}
}
.overlay {
if let errorMessage { Text(errorMessage).foregroundStyle(.red) }
}
.task {
do {
let stream = session.streamResponse(to: prompt, generating: TripIdeas.self)
for try await partial in stream {
partialResult = partial
}
} catch {
errorMessage = error.localizedDescription
}
}
}
| Decision | Rationale |
|----------|-----------|
| On-device execution | Privacy — no data leaves the device; works offline |
| 4,096 token limit | On-device model constraint; chunk large data across sessions |
| Snapshot streaming (not deltas) | Structured output friendly; each snapshot is a complete partial state |
| @Generable macro | Compile-time safety for structured generation; auto-generates PartiallyGenerated type |
| Single request per session | isResponding prevents concurrent requests; create multiple sessions if needed |
| response.content (not .output) | Correct API — always access results via .content property |
model.availability before creating a session — handle all unavailability casesinstructions to guide model behavior — they take priority over promptsisResponding before sending a new request — sessions handle one request at a timeresponse.content for results — not .output@Generable for structured output — stronger guarantees than parsing raw stringsGenerationOptions(temperature:) to tune creativity (higher = more creative)model.availability first.output instead of .content to access response data@Generable structured output would workAgent 工作流
创造有效技能指南。 当用户想创造出新的技能(或更新现有的技能),以专业知识,工作流程,或工具集成来扩展克洛德的能力时,应该使用这种技能.
Agent 工作流
使用ClawdHub CLI搜索,安装,更新并发布从taladhub.com的代理技能. 需要获取苍蝇上的新技能时使用,将安装的技能同步到最新版本或特定版本,或者发布 npm-instainddhub CLI 的新/更新的技能文件夹.
Agent 工作流
管弦乐团多代理团队,任务设定周期,交接协议,审查工作流程. 使用时间: (1)建立2+特派员队伍,具有不同专业,(2)确定任务路线和生命周期(收录框_ spec_建设_审查_完成),(3)在特派员之间制定交接协议,(4)建立审查和质量关口,(5)管理特派员之间的交流和文物共享.
Agent 工作流
Spec-first,TDD,子代理驱动的软件开发工作流程. 当:(1)构建任何新功能或应用——触发脑暴_计划_子代理执行回路,(2)调试出一个bug或测试失败——触发系统性的根起过程,(3)用户说"让我们构建","帮助我计划","我想添加X",或"这个被打破",(4)完成一个功…