scopes: Command for Claude Code

Run Tests

Execute tests with the configured test-runner agent.

Usage

/testing:run [test_target]

Where test_target can be:

• Empty (run all tests)
• Test file path
• Test pattern
• Test suite name

Quick Check

# Check if testing is configured
test -f .claude/testing-config.md || echo "❌ Testing not configured. Run /testing:prime first"

If test target provided, verify it exists:

# For file targets
test -f "$ARGUMENTS" || echo "⚠️ Test file not found: $ARGUMENTS"

Instructions

1. Determine Test Command

Based on testing-config.md and target:

• No arguments → Run full test suite from config
• File path → Run specific test file
• Pattern → Run tests matching pattern

2. Execute Tests

Use the test-runner agent from .claude/agents/test-runner.md:

Execute tests for: $ARGUMENTS (or "all" if empty)

Requirements:
- Run with verbose output for debugging
- No mocks - use real services
- Capture full output including stack traces
- If test fails, check test structure before assuming code issue

3. Monitor Execution

• Show test progress
• Capture stdout and stderr
• Note execution time

4. Report Results

Success:

✅ All tests passed ({count} tests in {time}s)

Failure:

❌ Test failures: {failed_count} of {total_count}

{test_name} - {file}:{line}
  Error: {error_message}
  Likely: {test issue | code issue}
  Fix: {suggestion}

Run with more detail: /testing:run {specific_test}

Mixed:

Tests complete: {passed} passed, {failed} failed, {skipped} skipped

Failed:
- {test_1}: {brief_reason}
- {test_2}: {brief_reason}

5. Cleanup

# Kill any hanging test processes
pkill -f "jest|mocha|pytest" 2>/dev/null || true

Error Handling

• Test command fails → "❌ Test execution failed: {error}. Check test framework is installed."
• Timeout → Kill process and report: "❌ Tests timed out after {time}s"
• No tests found → "❌ No tests found matching: $ARGUMENTS"