L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

image-loading

image-loading

Generates an image loading pipeline with memory/disk caching, deduplication, and a CachedAsyncImage SwiftUI view. Use when user wants image caching, lazy image loading, or a replacement for AsyncImage.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: image-loading description: Generates an image loading pipeline with memory/disk caching, deduplication, and a CachedAsyncImage SwiftUI view. Use when user wants image caching, lazy image loading, or a replacement for AsyncImage. allowed-tools: [Read, Write, Edit, Glob, Grep, Bash, AskUserQuestion] last_verified: 2026-07-16 review_by: 2027-06-22 os_version: iOS 27 / macOS 27

Image Loading Generator

Generate a production image loading pipeline with NSCache memory cache, LRU disk cache, request deduplication, image processing, and a drop-in CachedAsyncImage SwiftUI view.

When This Skill Activates

Use this skill when the user:

  • Asks to "add image caching" or "cache images"
  • Wants to "replace AsyncImage" or fix "AsyncImage has no cache"
  • Mentions "image loading pipeline" or "lazy image loading"
  • Asks about "image download" or "image prefetching"
  • Wants "thumbnail generation" or "image resizing"

Pre-Generation Checks

1. Project Context Detection

  • Check Swift version (requires Swift 5.9+)
  • Check deployment target (iOS 16+ / macOS 13+)
  • Check for @Observable support (iOS 17+ / macOS 14+)
  • Identify source file locations

2. Conflict Detection

Search for existing image loading:

Glob: **/*ImageCache*.swift, **/*ImageLoader*.swift, **/*ImagePipeline*.swift
Grep: "AsyncImage" or "UIImage" or "NSImage" or "ImageCache"

If third-party library found (Kingfisher, SDWebImage, Nuke):

  • Ask if user wants to replace or keep it
  • If keeping, don't generate — advise on best practices instead

3. Platform Detection

Determine if generating for iOS (UIImage) or macOS (NSImage) or both (cross-platform typealias).

Configuration Questions

Ask user via AskUserQuestion:

  1. Cache sizes?

    • Small (50 MB memory / 100 MB disk)
    • Medium (100 MB memory / 250 MB disk) — recommended
    • Large (200 MB memory / 500 MB disk)
  2. Image processing?

    • Resize to fit (downscale large images to save memory)
    • Thumbnail generation (create small thumbnails for lists)
    • None (cache original images only)
  3. Additional features? (multi-select)

    • Prefetching for collections (preload images for visible rows + buffer)
    • Placeholder and error images
    • Progress indicator during download
  4. Platform?

    • iOS only
    • macOS only
    • Cross-platform (iOS + macOS)

Generation Process

Step 1: Read Templates

Read image-loading-patterns.md for architecture guidance. Read templates.md for production Swift code.

Step 2: Create Core Files

Generate these files:

  1. ImageCache.swift — Protocol for cache interface
  2. MemoryImageCache.swift — NSCache-based with configurable size
  3. DiskImageCache.swift — FileManager LRU with expiration
  4. ImageDownloader.swift — Actor-based with deduplication + cancellation
  5. ImagePipeline.swift — Orchestrator (cache → download → process → store)

Step 3: Create UI Files

  1. CachedAsyncImage.swift — Drop-in SwiftUI view replacement

Step 4: Create Optional Files

Based on configuration:

  • ImageProcessor.swift — If resize or thumbnail selected
  • ImagePrefetcher.swift — If prefetching selected

Step 5: Determine File Location

Check project structure:

  • If Sources/ exists → Sources/ImageLoading/
  • If App/ exists → App/ImageLoading/
  • Otherwise → ImageLoading/

Output Format

After generation, provide:

Files Created

ImageLoading/
├── ImageCache.swift          # Protocol for cache interface
├── MemoryImageCache.swift    # NSCache-based memory cache
├── DiskImageCache.swift      # LRU disk cache with expiration
├── ImageDownloader.swift     # Actor-based downloader
├── ImagePipeline.swift       # Orchestrator
├── ImageProcessor.swift      # Resize, thumbnails (optional)
├── CachedAsyncImage.swift    # SwiftUI view
└── ImagePrefetcher.swift     # Collection prefetching (optional)

Integration Steps

Drop-in replacement for AsyncImage:

// Before (no caching)
AsyncImage(url: user.avatarURL) { image in
    image.resizable().aspectRatio(contentMode: .fill)
} placeholder: {
    ProgressView()
}

// After (with caching)
CachedAsyncImage(url: user.avatarURL) { image in
    image.resizable().aspectRatio(contentMode: .fill)
} placeholder: {
    ProgressView()
}

In a List:

List(users) { user in
    HStack {
        CachedAsyncImage(url: user.avatarURL) { image in
            image.resizable().frame(width: 44, height: 44).clipShape(Circle())
        } placeholder: {
            Circle().fill(Color.secondary.opacity(0.2)).frame(width: 44, height: 44)
        }
        Text(user.name)
    }
}

With prefetching:

struct UsersListView: View {
    let users: [User]
    @State private var prefetcher = ImagePrefetcher()

    var body: some View {
        List(users) { user in
            UserRow(user: user)
                .onAppear { prefetcher.startPrefetching(urls: nearbyURLs(for: user)) }
                .onDisappear { prefetcher.stopPrefetching(urls: [user.avatarURL]) }
        }
    }
}

With image processing:

CachedAsyncImage(
    url: photo.url,
    processing: .resize(targetSize: CGSize(width: 300, height: 300))
) { image in
    image.resizable()
} placeholder: {
    Color.secondary.opacity(0.2)
}

Testing

@Test
func cachedImageReturnedWithoutDownload() async throws {
    let cache = InMemoryImageCache()
    let downloader = MockImageDownloader()
    let pipeline = ImagePipeline(cache: cache, downloader: downloader)

    let testImage = PlatformImage.testImage
    await cache.store(testImage, for: testURL)

    let result = try await pipeline.image(for: testURL)
    #expect(result != nil)
    #expect(downloader.downloadCount == 0) // Cache hit
}

@Test
func deduplicatesConcurrentRequests() async throws {
    let downloader = MockImageDownloader(delay: .milliseconds(100))
    let pipeline = ImagePipeline(downloader: downloader)

    async let image1 = pipeline.image(for: testURL)
    async let image2 = pipeline.image(for: testURL)

    let results = try await [image1, image2]
    #expect(results.count == 2)
    #expect(downloader.downloadCount == 1) // Only one download
}

References

  • image-loading-patterns.md — Why not AsyncImage, NSCache config, LRU disk cache, deduplication
  • templates.md — All production Swift templates
  • Related: generators/http-cache — General HTTP response caching
  • Related: generators/pagination — Prefetch images in paginated lists
Skills enregistrées