By rshankras
SwiftShip — spec-driven /apple:* workflow for iOS/macOS apps: idea → validate → plan → build → review → App Store. Install apple-skills@indie-apple-stack alongside it.
Autonomously run plan→build→verify across remaining phases
Brainstorm app ideas tailored to your skills and interests
Quick bug fix with regression test and structured commit
Execute current plan using specialized agents
Systematic debugging with state tracking
Use this agent to check App Store Review Guidelines compliance before submission. Examples: <example> Context: User preparing for submission user: "Is my app ready for App Store?" assistant: "I'll use the app-store-reviewer agent to check Review Guidelines compliance." <commentary> Pre-submission checks trigger the app-store-reviewer agent. </commentary> </example> <example> Context: User got rejected user: "My app was rejected for guideline 2.1" assistant: "I'll use the app-store-reviewer agent to analyze and fix the rejection." <commentary> Rejection analysis triggers the agent. </commentary> </example>
Use this agent for CloudKit sync, iCloud integration, and data synchronization. Examples: <example> Context: User needs iCloud sync user: "Add iCloud sync to my app" assistant: "I'll use the cloudkit-expert agent to implement CloudKit synchronization." <commentary> iCloud/CloudKit work triggers the cloudkit-expert agent. </commentary> </example> <example> Context: User asks about sync conflicts user: "How do I handle sync conflicts?" assistant: "I'll use the cloudkit-expert agent to implement conflict resolution." <commentary> Sync-related questions trigger the agent. </commentary> </example>
Use this agent to review Human Interface Guidelines compliance for iOS/macOS apps. Examples: <example> Context: User wants HIG review user: "Review my app for HIG compliance" assistant: "I'll use the hig-reviewer agent to check Human Interface Guidelines compliance." <commentary> HIG review requests trigger the hig-reviewer agent. </commentary> </example> <example> Context: Proactive review after UI work user: [After building UI] assistant: "Let me use the hig-reviewer agent to verify HIG compliance." <commentary> After UI implementation, proactively review for HIG. </commentary> </example>
Use this agent for StoreKit 2 implementation, subscriptions, and in-app purchases. Examples: <example> Context: User needs to add purchases user: "Add in-app purchases to my app" assistant: "I'll use the storekit-expert agent to implement StoreKit 2 purchases." <commentary> Any IAP or StoreKit work triggers the storekit-expert agent. </commentary> </example> <example> Context: User asks about subscriptions user: "How do I handle subscription status?" assistant: "I'll use the storekit-expert agent to implement subscription handling." <commentary> Subscription questions trigger the agent. </commentary> </example>
Use this agent for Apple-platform build and review tasks that don't match a named specialist (swiftui-builder, storekit-expert, cloudkit-expert). It is the cost-pinned replacement for the built-in general-purpose agent — same breadth, but defaults to Sonnet instead of inheriting the session model (only an explicit per-spawn override from a command — a model="opus"-tagged plan task or a Critical verifier — runs it on anything else). Examples: <example> Context: /apple:build reaches a data-layer task user: "Task 3: Create SwiftData models for Deck and Card with relationships" assistant: "I'll use the swift-generalist agent with the macos/swiftdata-architecture skill to implement the models." <commentary> Data models map to no named specialist, so the generalist executes with the matched skill. </commentary> </example> <example> Context: /apple:review runs its parallel review pass user: "Review performance concerns for this Apple app" assistant: "I'll use the swift-generalist agent to scan for main-thread, memory, and launch-time issues." <commentary> The code-quality, performance, and security review passes run on the generalist in review mode (findings only, no edits). </commentary> </example>
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Spec-driven development for iOS & macOS apps, run entirely through Claude Code.
Every change is checked in CI: all 190+ skill references resolve against the skills library, documented counts match reality, every command is registered, and frontmatter is well-formed (scripts/validate.sh).
SwiftShip combines GSD's workflow methodology with deep Apple-platform expertise to walk you from "I have an app idea" all the way to "it's live on the App Store" — without losing context between sessions.
Imagine building an app with an AI assistant that, left alone, is brilliant but forgetful — it doesn't remember last week's decisions, and it doesn't automatically know all of Apple's rules.
SwiftShip is a project manager + a team of Apple specialists that sits on top of Claude Code. It gives you:
/apple:something) that take you from idea → design → build → test → ship..planning/ folder inside your project that records your spec, roadmap, decisions, and progress, so you can stop today and resume tomorrow exactly where you left off.Important: SwiftShip is not an app and not a program you compile. It's a set of carefully‑written instruction files (Markdown) that teach Claude Code how to behave like that project manager. There's nothing to build — you just install it and start typing commands.
Building an app is like building a house:
| House | SwiftShip |
|---|---|
| Blueprints & permits | the .planning/ files (spec, roadmap, plan) |
| The foreman you give orders to | the /apple:* commands you type |
| The plumber, electrician, roofer | the specialist agents the foreman calls in |
| The building‑code manuals on the shelf | the skills library (a separate companion project) |
┌─────────────────────────────────────────────────────────────────────────┐
│ COMMANDS (you type these) │
│ /apple:validate → /apple:new-app → /apple:roadmap → /apple:plan → ... │
└─────────────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────────────┐
│ PLANNING FILES (your project's memory, in .planning/) │
│ VALIDATION · APP · ROADMAP · STATE · PLAN · REVIEW · ASO · ... │
└─────────────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ AGENTS (specialists, called automatically) │
│ swiftui-builder · storekit-expert · hig-reviewer · ... │
└─────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ SKILLS (the "manuals" — from claude-code-apple-skills) │
│ ios/ · macos/ · generators/ · app-store/ · security/ · ... │
└─────────────────────────────────────────────────────────────┘
npx claudepluginhub rshankras/claude-code-apple-skills --plugin apple147 Apple platform development skills across 23 categories (iOS, macOS, watchOS, visionOS): code generators, product workflow, App Store/ASO, testing and TDD, design, performance, security, and more. Surfaced as 23 category skills that route to the full library.
Harness-native ECC plugin for engineering teams - 67 agents, 277 skills, 93 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
The Claude Code knowledge system — 380+ skills, 182+ agents, 100+ commands, 40 hooks, 32 rules, and workflows.
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Supergraph enforces a complete, evidence-based coding pipeline — scan → plan → TDD → fix → verify → review — grounded in real codebase analysis at every step. It combines AST dependency graphs, LSP-level code intelligence, and a structured skill chain so Claude never guesses about impact before making a change.