agent-skills

Integrate Apple Foundation Models into iOS apps — model selection, availability gates, prompting for the on-device model, guided generation, tool calling, Private Cloud Compute, token budgeting, locale handling, and model-version prompt updates. Use when building or reviewing Apple Intelligence / Foundation Models features in Swift or SwiftUI.

Index app content in Core Spotlight and build on-device “Ask” search with Foundation Models (CSSearchableItem, CSUserQuery, SpotlightSearchTool, search-then-summarize fallbacks, indexing progress, progressive results). Use when adding Spotlight indexing, reindex extensions, keyword/date search, or Apple Intelligence Ask over local notes/documents in Swift/SwiftUI apps.

Bootstrap a new or empty iOS/macOS app repo: XcodeGen project.yml, folder layout, ai-rules-ios (SwiftLint/SwiftFormat/Periphery), warnings-as-errors, and CI/preBuild quality wiring. Use when creating a new iOS project, starting a greenfield SwiftUI app, scaffolding project.yml from scratch, onboarding a repo that lacks .ai-rules, or when the user asks to set up lint/format/deadcode gates. For XcodeGen-only debugging of an existing spec, use xcodegen. For finishing work on an already-bootstrapped app, use ios-quality-gate.

Structure and refactor SwiftUI views for readability, reuse, and maintainability by extracting dedicated View structs, ViewModifiers, and layout helpers instead of bloated bodies or extension dumping grounds. Use when breaking down large views, improving screen clarity, creating reusable components, or reviewing SwiftUI architecture and view organization.

Run the shared ai-rules-ios quality plug-in (scripts/check.sh, format, Periphery deadcode --strict) before finishing iOS app work. Use when finishing features, refactors, removing APIs, cleanup, PRs, or Melvil/Gregor/Leio iOS changes that must stay lint-clean with zero unused code and no force unwraps.

Deprecated alias: use ios-bootstrap instead. Install ai-rules-ios and quality gates into an iOS app. Kept so older prompts that say bootstrap-ios-repo still resolve; prefer ios-bootstrap for new work.

Ship iOS apps to App Store Connect end-to-end — Xcode Cloud, ASC API metadata and screenshots, release manifests, TestFlight, and prepare-for-submission without auto-submit. Use when publishing an iOS app, setting up App Store automation, ASC API scripts, store screenshots, Xcode Cloud release workflows, or preparing a version for App Review.

Organize SwiftUI app repositories, folders, Swift packages, and architecture layers following Apple sample projects (Backyard Birds, Food Truck) and WWDC guidance. Use when scaffolding a new SwiftUI app, reorganizing features into modules, choosing MV vs ViewModel/Store layers, splitting SPM packages, structuring multiplatform targets (iOS/macOS/watchOS/widgets), wiring App/Scene entry points, or reviewing project-level architecture — not individual view bodies.

Author and debug XcodeGen project.yml specs — merge semantics, settings pitfalls, source filtering, dependency integration, multiplatform targets, schemes, and cache/CLI behavior. Use when editing an existing project.yml / project.yaml, fixing generated .xcodeproj issues, or when the user mentions XcodeGen, xcodegen generate, or spec-driven Xcode projects. For scaffolding a new iOS app (folders + ai-rules quality gates + warnings-as-errors), use ios-bootstrap instead.

Capture, compose, and upload iOS App Store screenshots via UI tests, xcresult export, Pillow marketing frames, and the App Store Connect API. Use when the user asks to generate store screenshots, refresh App Store screenshots, run capture-screenshots.sh / compose_store_assets.py / upload_store_assets.py, fix DemoMode screenshot fixtures, or submit screenshot sets to ASC — not for the full release/tag/TestFlight ship (use ios-app-store-release for that).

Create, update, release, and sync skills in the dbmrq/agent-skills repository (source of truth for personal Agent Skills). Use when the user asks to create a new skill, update an existing skill from a conversation or another repo, improve skill authoring practices, publish/release skills, or run install-all / gh skill sync — not when merely using a skill for its domain task.

Build iOS apps that look and behave like native Apple software by preferring highest-level SwiftUI components, system styles, semantic colors, and standard navigation structures. Use when creating or reviewing SwiftUI UI, choosing between custom vs built-in controls, styling buttons and cards, or when the user wants a native iOS look, HIG-aligned layouts, or out-of-the-box SwiftUI APIs instead of custom implementations.

Write Markdown implementation plans and checklists optimized for Ralph loops — one checkbox per agent run, autonomous ordering, integration tasks, verification criteria, and human-only gates at the end. Use when the user wants a Ralph loop plan, autonomous agent todo list, or says "use a ralph loop to build/create X" and needs the plan before running the loop.

Run and orchestrate Ralph loops — repeated fresh-context agent runs against a Markdown checklist until work is verified and complete. Covers loop theory, guardrails, per-run prompts, plan enrichment, and writing ralph-loop shell scripts. Use when the user mentions Ralph loops, ralph-loop.sh, autonomous agent iteration, or wants to run agents repeatedly against a plan checklist.