---

name: test-runner description: Execute VitalArc tests with various options. Required quality gate at session end. Supports quick, coverage, and affected-only modes. context: fork agent: Bash allowed-tools: Bash, Read, Grep, Glob, TaskCreate, TaskUpdate, TaskList argument-hint: [--quick] [--coverage] [--affected] [--filter=pattern]

Test Runner

Executes VitalArc tests as a required quality gate. Workstation only - requires Xcode build capability.

Execution: Runs in forked context with Bash agent.

IMPORTANT: When invoked without arguments, execute immediately in Full mode (run all tests). Never ask for clarification - use defaults and produce results.

When to Use

• Required: At session end (blocking gate after build passes)
• Before committing significant changes
• After implementing new features
• When user asks to "run tests", "verify tests", or "check tests"

Test Modes

Mode	Flag	Description	Use Case
Full	(default)	Run all tests	Complete validation
Quick	--quick	Run fast unit tests only	Rapid feedback
Coverage	--coverage	Generate coverage report	Before PR
Affected	--affected	Tests for changed files only	Incremental check

Implementation

Full Test Run (Default)

# Run all tests
xcodebuild test \
    -scheme VitalArc \
    -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \
    -resultBundlePath /tmp/test-results.xcresult \
    2>&1 | xcpretty --color

# Check exit code
if [ $? -eq 0 ]; then
    echo "ALL TESTS PASSED"
else
    echo "TESTS FAILED"
    exit 1
fi

Quick Mode (--quick)

Runs only fast unit tests, skipping UI and integration tests:

# Run only unit tests (exclude UI tests)
xcodebuild test \
    -scheme VitalArc \
    -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \
    -only-testing:VitalArcTests \
    2>&1 | grep -E "(Test Case|passed|failed|error:)"

Coverage Mode (--coverage)

Generates code coverage report:

# Run with coverage enabled
xcodebuild test \
    -scheme VitalArc \
    -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \
    -enableCodeCoverage YES \
    -resultBundlePath /tmp/coverage-results.xcresult \
    2>&1 | xcpretty --color

# Extract coverage report
xcrun xccov view --report /tmp/coverage-results.xcresult

Affected Mode (--affected)

Only tests related to changed files:

# Get changed files
CHANGED_FILES=$(git diff --name-only HEAD~1 -- '*.swift')

# Find related test files
for file in $CHANGED_FILES; do
    base=$(basename "$file" .swift)
    # Look for corresponding test file
    TEST_FILE=$(find VitalArcTests -name "${base}Tests.swift" 2>/dev/null)
    if [ -n "$TEST_FILE" ]; then
        TESTS_TO_RUN="$TESTS_TO_RUN -only-testing:VitalArcTests/${base}Tests"
    fi
done

# Run only affected tests
xcodebuild test \
    -scheme VitalArc \
    -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \
    $TESTS_TO_RUN \
    2>&1 | xcpretty --color

Filter Mode (--filter=pattern)

Run tests matching a pattern:

PATTERN="$1"

# Run tests matching pattern
xcodebuild test \
    -scheme VitalArc \
    -destination 'platform=iOS Simulator,name=iPhone 17 Pro' \
    -only-testing:VitalArcTests \
    2>&1 | grep -E "(Test Case.*$PATTERN|passed|failed)"

Output Format

Success Report

## Test Results

**Status**: PASSED
**Mode**: Full test run
**Duration**: 45 seconds

### Summary
| Category | Tests | Passed | Failed |
|----------|-------|--------|--------|
| Unit Tests | 42 | 42 | 0 |
| ViewModel Tests | 18 | 18 | 0 |
| Integration Tests | 8 | 8 | 0 |
| **Total** | **68** | **68** | **0** |

### Test Suites
- HealthKitTests: 12 tests, 12 passed
- NutritionTests: 15 tests, 15 passed
- ProfileTests: 10 tests, 10 passed
- TemplateTests: 8 tests, 8 passed
- BMITests: 5 tests, 5 passed
- WorkoutTests: 18 tests, 18 passed

All tests passed. Ready for commit.

Failure Report

## Test Results

**Status**: FAILED
**Mode**: Full test run
**Duration**: 52 seconds

### Summary
| Category | Tests | Passed | Failed |
|----------|-------|--------|--------|
| Unit Tests | 42 | 40 | 2 |
| ViewModel Tests | 18 | 18 | 0 |
| Integration Tests | 8 | 8 | 0 |
| **Total** | **68** | **66** | **2** |

### Failed Tests

#### NutritionTests.swift

**testCalorieCalculation_withZeroPortions_returnsZero**

XCTAssertEqual failed: ("100.0") is not equal to ("0.0") Location: NutritionTests.swift:45

**testMealTotals_withEmptyMeal_returnsDefaults**

XCTAssertNil failed: "Optional(Meal)" - Expected nil Location: NutritionTests.swift:78

### Suggested Fixes

1. **testCalorieCalculation**: Check portion handling in `calculateCalories()`
2. **testMealTotals**: Verify empty state initialization in `Meal.swift`

---

**BLOCKED**: Fix failing tests before session end.

Coverage Report

## Test Coverage Report

**Overall Coverage**: 68%

### Coverage by File

| File | Lines | Covered | % |
|------|-------|---------|---|
| ProfileViewModel.swift | 150 | 135 | 90% |
| NutritionUseCase.swift | 89 | 78 | 88% |
| WorkoutManager.swift | 210 | 168 | 80% |
| HealthKitManager.swift | 180 | 108 | 60% |
| NotificationService.swift | 120 | 48 | 40% |

### Uncovered Areas

**HealthKitManager.swift** (60% coverage)
- Lines 45-67: Error handling paths
- Lines 120-145: Background refresh logic

**NotificationService.swift** (40% coverage)
- Lines 30-55: Notification scheduling
- Lines 78-95: Permission request handling

### Recommendations
1. Add tests for HealthKit error scenarios
2. Add tests for notification scheduling
3. Consider mock framework for better isolation

Integration with Session End

As Required Gate

In vitalarc-end-workstation, test execution is required after build passes:

PHASE 1: Build validation (BLOCKING)
PHASE 2: Test execution (BLOCKING - after build)
PHASE 3: Parallel quality checks (after tests)
PHASE 4: Commit generation

// In vitalarc-end-workstation Phase 2:
TaskCreate({
  subject: "Run test suite",
  description: `Execute tests as required quality gate:
    1. Run: /test-runner (full mode)
    2. If FAILED: Block session end, report failures
    3. If PASSED: Proceed to quality checks`,
  activeForm: "Running tests",
  addBlockedBy: ["task-build-id"]  // After build passes
})
// Returns: task-test-id

// Phase 3 tasks blocked by test-id
TaskCreate({
  subject: "Design system scan",
  ...
  addBlockedBy: ["task-test-id"]  // After tests pass
})

Test Failure Handling

If tests fail, session end is blocked:

═══════════════════════════════════════════════════════════════
       SESSION END BLOCKED - TESTS FAILED
═══════════════════════════════════════════════════════════════
Failed Tests: 2

1. NutritionTests.testCalorieCalculation
2. NutritionTests.testMealTotals

Fix failing tests, then re-run /vitalarc-end-workstation
═══════════════════════════════════════════════════════════════

Error Handling

Simulator Not Available

## Test Run Failed

**Error**: iOS Simulator not available

The specified simulator "iPhone 17 Pro" is not available.

**Available simulators**:
```bash
xcrun simctl list devices available

Update test destination or create required simulator.

### Build Required

```markdown
## Test Run Failed

**Error**: Build required before testing

Tests cannot run without a successful build.

Run `/build-validator` first, then retry tests.

Timeout

## Test Run Warning

**Warning**: Tests taking longer than expected

Running for: 5 minutes (expected: ~2 minutes)

Consider:
- Running `--quick` mode for faster feedback
- Running `--affected` to test only changed code
- Checking for hanging tests or infinite loops

Best Practices

1. Run quick tests frequently during development
2. Run full tests before commits to catch regressions
3. Run coverage before PRs to ensure adequate coverage
4. Run affected tests for rapid iteration during feature work
5. Never skip tests - they're a required gate for a reason