# QA Test Creation > Test coverage check and creation workflow. Ensures tests exist for all features. Creates unit tests when missing following acceptance criteria. Use during validation before code review. Use when this capability is needed. - Skill: `tomevault-io/qa-test-creation` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/qa-test-creation` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/qa-test-creation/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Productivity - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/qa-test-creation --- # Test Coverage Check and Creation Workflow > "No feature passes QA without tests. If tests don't exist, we create them." ## When to Use This Skill Use when: - Validating a new feature (status: "awaiting_qa") - PRD item has acceptance criteria but no tests exist - Source files are modified but corresponding test files are missing **CRITICAL:** NEVER CREATE FAKE OR TRIVIAL TESTS. If a test cannot be created with meaningful assertions based on the specs/gdd requirements, DO NOT CREATE A TEST. Instead, report the issue to PM and document the gap in test coverage in the task comments and close the task as completed with observations. ## Test Coverage Guidelines **⚠️ CRITICAL: Focus on unit tests for code quality verification.** QA validates implementation through: - Code review for quality, anti-patterns, and maintainability - Unit tests for component behavior, services, and utilities - Specification validation against acceptance criteria **Avoid browser/E2E testing** - AI agents do not reliably handle browser automation, which delays validation. QA must be smart enough to validate code quality through static analysis and unit testing. ### Unit Test Focus | Component Type | Unit Test Coverage | Spec Validation | | --------------- | ------------------- | ----------------- | | Gameplay mechanic | YES (systems, components) | YES | | UI Component | YES (rendering, behavior) | YES | | Visual effect | YES (shader/material params) | YES | | Service/Utility | YES (pure functions) | MAYBE | | Asset (model/texture) | N/A | YES (visual inspection) | **CRITICAL:** NEVER CREATE FAKE OR TRIVIAL TESTS. If a meaningful unit test cannot be created based on implementation specs, report the gap to PM and document as observation in task comments. ## Quick Start ```bash # 1. Get the task ID from PRD # 2. Check test coverage for modified files # 3. If coverage incomplete, invoke test-creator # 4. Verify tests pass before proceeding ``` ## Test File Locations ### Unit Tests (Vitest) - **Pattern**: Mirror `src/` structure in `src/tests/` - **Example**: `src/components/game/player/index.ts` → `src/tests/components/game/player/index.test.ts` - **Check**: For each source file, check if corresponding `src/tests/.../{name}.test.ts` exists ## Coverage Check Procedure ### Step 1: Get Task Information ```bash # Read PRD for current task cat prd.json | jq '.items[] | select(.id == "{taskId}")' # Extract: # - acceptance criteria # - modified files (from git diff or PRD) # - feature type (component, service, gameplay, etc.) ``` ### Step 2: Identify Files to Check ```bash # Get list of modified source files git diff --name-only HEAD~5 HEAD | grep '^src/' # Or read from task context # Common patterns: # - src/components/**/*.tsx # - src/services/**/*.ts # - src/stores/**/*.ts # - src/utils/**/*.ts # - src/ecs/**/*.ts ``` ### Step 3: Check Unit Test Coverage For each source file: ```bash # Convert source path to test path SOURCE="src/components/game/player/index.ts" TEST="src/tests/components/game/player/index.test.ts" # Check if test exists if [ -f "$TEST" ]; then echo "Test exists: $TEST" else echo "MISSING TEST: $TEST" fi ``` ### Step 4: Determine Required Tests | Source File Pattern | Requires Unit Test? | | ------------------- | ------------------- | | `src/components/**/*.tsx` | Yes | | `src/services/**/*.ts` | Yes | | `src/stores/**/*.ts` | Yes | | `src/utils/**/*.ts` | Yes | | `src/ecs/**/*.ts` | Yes | | `src/audio/**/*.ts` | Yes | | `src/materials/**/*.ts` | No (visual asset) | | `src/shaders/**/*.ts` | No (visual asset) | ### Step 6: Invoke Test Creator **If tests are missing, invoke the test-creator sub-agent:** ```javascript Task({ subagent_type: "test-creator", description: "Create tests for {feature-name}", prompt: ` Create tests for the following task: Task ID: {taskId} Title: {title} Acceptance Criteria: {list of acceptance criteria} Modified Files: {list of modified files} GDD Specs: {relevant GDD sections} Create: 1. Unit tests in src/tests/ mirroring src structure 2. E2E tests in tests/e2e/ with {feature}-suite.spec.ts naming Ensure all acceptance criteria have test coverage. ` }) ``` ### Step 6: Execute Tests and Verify After tests are created (or if they already existed): ```bash # Run unit tests npm run test # Verify tests pass # Check exit codes and output ``` ### Step 8: Test Failure Analysis When tests fail, determine the root cause. > **See `qa-workflow` skill for the Test Failure Decision Tree and how to distinguish test vs game code issues.** ### Step 8: Fix Test Code (If Applicable) If issue is in test code, QA may fix: ```bash # Edit test file directly (QA has permission for test files) Edit src/tests/{path}/{feature}.test.ts # Re-run specific test npm run test -- src/tests/{path}/{feature}.test.ts # Verify fix git diff src/tests/{path}/{feature}.test.ts ``` ### Step 9: Create Bug Report (If Game Code Issue) If issue is in game code, use `qa-bug-reporting` skill: ```markdown ## Bug Report: {TASK_ID} - Test Failure **Severity**: High **Category**: Test / Runtime ### Summary Unit test "{test_name}" failed due to game code not meeting acceptance criteria. ### Test That Failed - File: src/tests/{path}/{feature}.test.ts - Test: "{test_name}" - Error: {error_message} ### Acceptance Criteria Not Met - {criterion from test plan} ### Expected vs Actual - Expected: {what test expects} - Actual: {what actually happened} ### Test Output \`\`\` {npm run test output} \`\`\` ``` ## Coverage Report Template After checking coverage, create a report: ```markdown ## Test Coverage Report for {taskId} ### Unit Tests | Source File | Test File | Status | | ------------ | --------- | ------ | | src/components/game/player/index.ts | src/tests/components/game/player/index.test.ts | ✅ EXISTS / ❌ MISSING | | src/services/ShootingService.ts | src/tests/services/ShootingService.test.ts | ✅ EXISTS / ❌ MISSING | ### Coverage Summary - Unit test coverage: X% - Action required: CREATE TESTS / NO ACTION ``` ## Decision Tree ``` ┌─────────────────┐ │ Start Check │ └────────┬────────┘ │ ▼ ┌─────────────────┐ │ Any source │ │ files modified? │ └────────┬────────┘ │ ┌────────────┴────────────┐ │ NO │ YES ▼ ▼ ┌──────────┐ ┌──────────────────┐ │ SKIP │ │ Check each source │ │ (assets │ │ file for test │ │ only) │ └─────────┬────────┘ └──────────┘ │ ┌────────────┴────────────┐ │ │ Test exists Test missing │ │ ▼ ▼ ┌──────────┐ ┌──────────────┐ │ Mark OK │ │ Invoke test- │ └──────────┘ │ creator │ └──────────────┘ ``` ## Integration with QA Workflow **Add this step between "Task Research" and "Code Review":** ``` 6. TASK RESEARCH (MANDATORY) ↓ 7. TEST COVERAGE CHECK (NEW) - Check unit test coverage - Create missing tests via test-creator ↓ 8. Run validation: code review → type-check → lint → test → build ``` ## Test Creation Triggers **Invoke test-creator when:** 1. **New feature implementation** - No tests exist for new code 2. **Missing unit tests** - Source file exists but no `.test.ts` in `src/tests/` 3. **Acceptance criteria untested** - Criterion has no corresponding test case **Do NOT invoke test-creator when:** 1. Tests already exist and pass 2. Only configuration/doc changes 3. Asset file changes (models, textures) 4. Test refactoring (updating existing tests) ## Verification After Test Creation After test-creator completes: ```bash # 1. Verify tests were created git status git diff --cached # 2. Run unit tests npm run test # 3. Verify coverage npm run test -- --coverage # 4. Proceed with validation workflow ``` ## Examples ### Example 1: Player Movement Feature **PRD Item:** ```json { "id": "feat-movement-001", "title": "WASD Movement System", "acceptanceCriteria": [ "Player moves forward when W is pressed", "Player moves left when A is pressed", "Player moves backward when S is pressed", "Player moves right when D is pressed" ], "files": [ "src/components/game/player/index.tsx", "src/ecs/systems/MovementSystem.ts" ] } ``` **Coverage Check:** ```bash # Unit tests src/tests/components/game/player/index.test.ts ❌ MISSING src/tests/ecs/systems/MovementSystem.test.ts ❌ MISSING ``` **Action:** Invoke test-creator to create missing tests ### Example 2: UI Component Only **PRD Item:** ```json { "id": "feat-ui-001", "title": "Health Bar Component", "acceptanceCriteria": [ "Displays current health", "Updates when health changes", "Shows correct color based on health percentage" ], "files": [ "src/components/ui/HealthBar.tsx" ] } ``` **Coverage Check:** ```bash # Unit tests src/tests/components/ui/HealthBar.test.ts ❌ MISSING ``` **Action:** Invoke test-creator to create missing unit tests ## Best Practices 1. **Be thorough** - Every acceptance criterion needs test coverage 2. **Test behavior, not implementation** - Focus on what the feature does 3. **Use proper test structure** - AAA pattern for unit tests 4. **Ensure tests are independent** - No test dependencies 5. **Make tests fast** - Unit tests should run in milliseconds 6. **Use descriptive names** - Test names should explain what is being tested ## References - [qa-unit-test-creation](./.claude/skills/qa-unit-test-creation/SKILL.md) - Unit test patterns - [test-creator agent](./.claude/agents/test-creator.md) - Test creation sub-agent --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/feliperyba) — claim your Tome and manage your conversions.