Ios Test logo

Ios Test

Community
yusufkaran
ios-test

Build, launch, and visually test iOS/SwiftUI apps in the Simulator using computer use. Automated screen navigation, crash log analysis, state testing (empty/error/loading), and memory leak detection. Use when you need to test an iOS app, run it in the Simulator, check for crashes, or verify UI flows.

Overview

Publisheryusufkaran
Repositoryswiftui-autotest-skill
Skill nameios-test
Stars
23
Forks
0
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by yusufkaran on GitHub. Read the source before you install it.

Installation

Install the Ios Test AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/yusufkaran/swiftui-autotest-skill/tree/main/skills/ios-test
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/yusufkaran/swiftui-autotest-skill.git /tmp/swiftui-autotest-skill
mkdir -p .claude/skills
cp -r /tmp/swiftui-autotest-skill/skills/ios-test .claude/skills/ios-test
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ios Test in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ios Test on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ios Test is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

iOS Test Skill

Overview

Use this skill to build an iOS/SwiftUI application, launch it in the Simulator, and visually test it using computer use. The skill navigates through every screen, screenshots each state, checks crash logs, and produces a structured test report — without writing a single line of test code.

Options

  • --flow=<name>: Test a specific user flow (e.g., onboarding, login, checkout)
  • --screen=<ViewName>: Test only a specific screen
  • --device=<name>: Choose a Simulator device (e.g., "iPhone 16", "iPhone SE")
  • --scheme=<name>: Specify the Xcode scheme
  • --states: Test empty, error, and loading states via launch arguments
  • --screenshot-all: Screenshot every step
  • --performance: Measure RAM per screen and check for memory leaks (skipped by default)

Workflow

1) Project Discovery

  • Find .xcodeproj or .xcworkspace in the working directory
    • Prefer .xcworkspace if it exists (CocoaPods/SPM workspace)
    • If multiple exist, ask the user which one to use
    • If none found, stop with an error
  • List available schemes:
    bash
    xcodebuild -list -workspace MyApp.xcworkspace
    • Use --scheme if provided
    • If only one scheme exists, use it automatically
    • If multiple, ask the user

2) Simulator Selection (BEFORE build)

IMPORTANT: Simulator must be selected BEFORE building. The build command uses the selected device name.

  • If --device is provided: find that device via xcrun simctl list devices --json
  • If --device is not provided, check booted simulators:
    bash
    xcrun simctl list devices booted --json
    • No booted simulator: list available devices with xcrun simctl list devices available --json, suggest the best iPhone match, ask for confirmation, then boot it
    • One booted simulator: use it directly, no questions
    • Multiple booted: list them and ask the user which one to use

3) Accessibility Check (MANDATORY — DO NOT SKIP)

This step is MANDATORY. Runs BEFORE build. NEVER skip this step.

  1. Scan SwiftUI files in the project — look for struct ... : View patterns in *.swift files
  2. Count interactive elements: Button, TextField, SecureField, Toggle, Slider, Picker, DatePicker, NavigationLink, Image, .onTapGesture
  3. Check how many have .accessibilityIdentifier()
  4. ALWAYS ask the user (this question cannot be skipped):
Accessibility Scan Result:
   Total interactive elements: XX
   With identifier: XX
   Missing identifier: XX

Identifiers help me find elements more reliably during testing.

Run /add-accessibility to add identifiers now?
  → Yes: adds identifiers, then builds and continues to testing
  → No: builds directly and uses coordinate-based testing (slower and more fragile)
  1. WAIT for the user's response. Do not proceed without an answer.
  2. If yes → run the full /add-accessibility workflow (scan, generate, apply), then continue to step 4 (Build)
  3. If no → continue to step 4 (Build) directly

4) Build

  • Build the selected scheme using the device chosen in step 2:
    bash
    xcodebuild build \
      -workspace MyApp.xcworkspace \
      -scheme MyApp \
      -destination 'platform=iOS Simulator,name=<SELECTED_DEVICE>' \
      -derivedDataPath ./DerivedData \
      2>&1
    NEVER hardcode a device name. Always use the device selected in step 2.
  • If build fails: analyze errors, suggest fixes, and STOP — never proceed to testing with a failed build
  • Report build warnings (but don't stop)

5) Install & Launch

  • Find the built .app file:
    bash
    find ./DerivedData -name "*.app" -path "*/Debug-iphonesimulator/*" | head -1
  • Get the bundle identifier:
    bash
    plutil -p /path/to/MyApp.app/Info.plist | grep CFBundleIdentifier
  • Install and launch:
    bash
    xcrun simctl install booted /path/to/MyApp.app
    xcrun simctl launch booted <BUNDLE_ID>

6) Visual Testing with Computer Use

IMPORTANT: This phase requires the computer-use MCP server to be enabled.

If computer use is not enabled, tell the user:

Computer use is not enabled. Required for visual testing.
Run /mcp and enable the computer-use server.
Default test (no arguments)

Discover and test all main screens:

  1. TabView present → tap each tab, inspect each screen
  2. NavigationStack present → tap each navigation link, go back
  3. On each screen check:
    • Layout renders correctly (no overflow, overlapping, or empty areas)
    • Buttons are tappable
    • Scroll works and content is visible
    • Text is readable (not too small, not truncated)
  4. Screenshot each screen
With --flow=<name>

Test the specified user flow:

  • onboarding: swipe through onboarding screens, complete all steps
  • login: fill email/password fields (test@example.com / Test1234), tap login, verify next screen
  • signup: fill the registration form, tap signup
  • checkout: navigate to cart, complete the payment flow
  • settings: open settings, try each toggle/slider, go back
  • Other flows: ask the user "What steps should I follow for this flow?"
With --screen=<ViewName>

Find and test only the specified screen. Navigate to it if needed.

7) State Testing (--states only)

Test different app states using launch arguments.

  1. Check if the app supports state overrides via CommandLine.arguments:

    • Search SwiftUI files for --show-empty-state, --show-error-state, --show-loading-state
    • Search for @AppStorage or UserDefaults debug flags
  2. If not found, ask the user:

    No state test support found. Would you like me to add launch argument
    handling to your @main App struct?
    
    A #if DEBUG block will be added supporting:
    - --show-empty-state
    - --show-error-state
    - --show-loading-state
    
    Only active in DEBUG builds. Add it?
  3. If available, relaunch the app for each state:

    bash
    xcrun simctl terminate booted com.example.MyApp
    xcrun simctl launch booted com.example.MyApp --show-empty-state

    For each state verify:

    • Empty state: shows a meaningful message ("No content yet")
    • Error state: clear error message with a retry button
    • Loading state: loading indicator visible, UI not frozen
    • Screenshot each state

8) Crash Log Analysis

Check for crashes during and after testing:

bash
xcrun simctl spawn booted log show --predicate 'process == "MyApp" AND messageType == 21' --last 5m

If a crash is found:

  • Analyze the stack trace
  • Identify which screen / action triggered it
  • Find the relevant source code and report the likely cause

9) Performance Analysis (--performance only)

ONLY run when --performance is provided. SKIP in default tests.

Measure RAM usage per screen and check for memory leaks.

  1. Find the app's PID:

    bash
    xcrun simctl spawn booted launchctl list | grep <BundleID>
  2. Take a baseline measurement on launch:

    bash
    footprint -all <PID>
  3. On each screen transition:

    • Navigate to the screen (via computer use)
    • Wait 2 seconds (let rendering complete)
    • Run footprint -all <PID>, record RAM
    • Calculate diff from previous screen
  4. Check for leaks on back navigation:

    • Enter a screen → measure RAM
    • Go back → measure RAM
    • If delta > 5MB → flag as potential leak
  5. Run a full leak scan at the end:

    bash
    leaks <PID>

    Report: leak count, object types, stack traces

  6. Performance report format:

    Performance Report
    ━━━━━━━━━━━━━━━━━━━━━━━━━
    Baseline (launch):       42 MB
    
    Screen Transitions:
      HomeView:              45 MB (+3 MB)
      SettingsView:          52 MB (+7 MB)
      ProfileView:           68 MB (+16 MB) ⚠️
      → HomeView (back):     61 MB (-7 MB, 16 MB not released) ⚠️
    
    Memory Leaks:            2 found
      - SettingsViewModel: 1 leak (closure retain cycle?)
      - ImageCache: 1 leak
    
    ⚠️ High RAM: ProfileView (+16 MB jump)
    ⚠️ Potential Leak: 16 MB not released on return to HomeView

10) Test Report

Show a summary report when testing is complete:

iOS Test Report
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
App:         MyApp (com.example.MyApp)
Device:      iPhone 16 (iOS 18.2)
Scheme:      MyApp
Duration:    2m 34s

Screens Tested: 8
  ✅ HomeView - OK
  ✅ ProfileView - OK
  ⚠️ SettingsView - Toggle "notifications" not tappable
  ✅ OnboardingStep1 - OK
  ✅ OnboardingStep2 - OK
  ✅ OnboardingStep3 - OK
  ❌ CheckoutView - Crash (force unwrap on nil)
  ✅ SearchView - OK

Screenshots: 12 captured

Crashes: 1
  CheckoutView.swift:42 - Force unwrap on nil optional

UI Issues: 1
  SettingsView - notifications toggle not accessible

State Test Results:
  ✅ Empty State - OK
  ⚠️ Error State - Missing retry button
  ✅ Loading State - OK

Important Rules

  1. NEVER proceed to testing if the build fails
  2. Cannot do visual testing without computer use — tell the user to enable it
  3. Don't ask unnecessary questions about Simulator selection — use the one that's booted
  4. Report every error and warning but NEVER modify code without user approval
  5. Don't delete or modify the app's data during testing
  6. Don't suggest adding launch arguments unless --states is provided
  7. When a crash is detected, show the stack trace alongside the relevant source code
  8. Save screenshots with meaningful names (e.g., test-home-screen.png)
  9. Do NOT close the Simulator after testing — the user may want to continue
  10. Provide brief status messages at each phase ("Building...", "Launching Simulator...", etc.)

Frequently asked questions

What does the Ios Test AI skill do?

Build, launch, and visually test iOS/SwiftUI apps in the Simulator using computer use. Automated screen navigation, crash log analysis, state testing (empty/error/loading), and memory leak detection. Use when you need to test an iOS app, run it in the Simulator, check for crashes, or verify UI flows.

Why use Ios Test on TypingMind?

Because you install it once and use it with any model. Ios Test is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ios Test in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/yusufkaran/swiftui-autotest-skill/tree/main/skills/ios-test. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Ios Test?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ios Test?

As many as you like. As long as a model supports skills, you can use Ios Test with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ios Test AI skill free?

Yes. It is published on GitHub by yusufkaran under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇