AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Ios Test

skill-yusufkaran-swiftui-autotest-skill-ios-test · by yusufkaran

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.

No reviews yet
0 installs
6 views
0.0% view→install

Install

$ agentstack add skill-yusufkaran-swiftui-autotest-skill-ios-test

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-yusufkaran-swiftui-autotest-skill-ios-test)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Ios Test? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

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=: Test a specific user flow (e.g., onboarding, login, checkout)
  • --screen=: Test only a specific screen
  • --device=: Choose a Simulator device (e.g., "iPhone 16", "iPhone SE")
  • --scheme=: 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=' \ -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 ``

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)
  1. Screenshot each screen
With --flow=

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=

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
  1. 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? ```

  1. 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:

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 ``

  1. Take a baseline measurement on launch:

``bash footprint -all ``

  1. On each screen transition:
  • Navigate to the screen (via computer use)
  • Wait 2 seconds (let rendering complete)
  • Run footprint -all , record RAM
  • Calculate diff from previous screen
  1. Check for leaks on back navigation:
  • Enter a screen → measure RAM
  • Go back → measure RAM
  • If delta > 5MB → flag as potential leak
  1. Run a full leak scan at the end:

``bash leaks `` Report: leak count, object types, stack traces

  1. 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.)

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.