testing

name	testing
description	Write and run tests. Tests are REQUIRED. All must pass.

/testing [level]

Write and RUN tests. Tests are REQUIRED, not optional.

No arguments? Describe this skill and stop. Do not execute.

First: Activate Workflow

mkdir -p .claude && echo '{"skill":"test","started":"'$(date -Iseconds)'"}' > .claude/active-workflow.json

Craft Standards (MANDATORY)

Write tests a master craftsperson would be proud of.

Tests must look like they were written by a skilled human engineer, not generated by AI.

AI Antipatterns in Tests to REJECT

• Testing implementation details instead of behavior
• Excessive mocking that makes tests brittle
• Mocking the system under test — if you're testing storage and you mock storage, you're testing nothing. Mocks replace dependencies, never the thing being tested.
• Tests that pass even when code is broken (useless assertions)
• Copy-paste test code (extract helpers)
• Testing obvious things (constructors, getters)
• Overly verbose arrange/act/assert with boilerplate

Human Craft in Tests

• Tests that document intended behavior
• Clear test names that read as specifications
• Focused tests - one concept per test
• Tests that fail meaningfully when code breaks
• Minimal setup - only what's needed
• Tests others can read and understand

Test: Would you trust this test suite to catch real bugs? If not, improve it.

---

⚠️ STRICT REQUIREMENTS - NO JUDGMENT CALLS

You MUST write and run tests. Not "consider testing" - WRITE TESTS.

1. MINIMUM COVERAGE - At least one test per public function
2. HAPPY PATH - Test the expected behavior works
3. ERROR CASES - Test that errors are handled correctly
4. EDGE CASES - Test boundary conditions
5. NON-HAPPY-PATH - Test failure and recovery scenarios (see below)
6. MUST RUN - Execute tests with npm test/vitest/jest and verify they pass
7. ALL MUST PASS - Zero failing tests allowed

Non-Happy-Path Categories (MANDATORY)

For each category that applies to the target code, write at least one test:

• Corrupted data recovery — What happens when persisted files contain invalid JSON, truncated data, wrong version, or zero bytes? Test that the code fails gracefully with a clear error, not a stack trace or silent corruption.
• Lock contention — If the code uses file locks or mutexes, test what happens when the lock is already held (timeout, retry, clear error message).
• Interrupted writes — If the code writes files, test what happens when the write is interrupted (temp file left behind, partial write). Verify cleanup occurs.
• Symlink / path escape — If the code validates paths, test that symlinks pointing outside the allowed directory are rejected.
• Resource exhaustion — If the code reads unbounded input (stdin, files), test that size limits are enforced.

Skip categories that genuinely don't apply (e.g., no file I/O means no corrupted data test). But document which categories you skipped and why.

FORBIDDEN (Phase will FAIL if detected):

• Saying "testing would be beneficial" without writing tests
• Writing tests that don't run
• Leaving failing tests
• Skipping error case testing
• Writing trivial tests that don't verify behavior
• Not running the tests
• Mocking the module under test (mock dependencies, not the subject)

Levels

• /testing unit - Unit tests only
• /testing integration - Integration tests only
• /testing e2e - E2E tests only
• /testing all or /testing - All appropriate levels

Process

Step 0: Load Expert Guidance

Before starting, read these canon skills and apply their principles throughout:

Always load:

1. canon/testing/test-doubles/SKILL.md
2. canon/testing/test-strategy/SKILL.md

Auto-detect domain canon (check files, load matches):

Check	If found, also read
*.ts or *.js files in target	canon/javascript/typescript/SUMMARY.md, canon/javascript/js-safety/SUMMARY.md, canon/javascript/js-perf/SUMMARY.md, canon/javascript/js-internals/SUMMARY.md, canon/javascript/functional/SUMMARY.md
angular.json in project root	canon/angular/angular-arch/SUMMARY.md, canon/angular/angular-core/SUMMARY.md, canon/angular/angular-perf/SUMMARY.md, canon/angular/rxjs/SUMMARY.md
package.json contains "react"	canon/javascript/react-state/SUMMARY.md, canon/javascript/react-test/SUMMARY.md, canon/javascript/reactivity/SUMMARY.md
pom.xml or build.gradle in project	canon/java/SUMMARY.md
*.py files in target	canon/python/python-advanced/SUMMARY.md, canon/python/python-idioms/SUMMARY.md, canon/python/python-patterns/SUMMARY.md, canon/python/python-protocols/SUMMARY.md
*.cs files or *.csproj in project	canon/csharp/csharp-depth/SUMMARY.md, canon/csharp/type-systems/SUMMARY.md, canon/csharp/async/SUMMARY.md
.tsx, .jsx, or HTML template files	canon/ui-ux/components/SUMMARY.md, canon/ui-ux/usability/SUMMARY.md, canon/ui-ux/tokens/SUMMARY.md
Auth, tokens, secrets, encryption	canon/security/security-mindset/SUMMARY.md, canon/security/owasp/SUMMARY.md, canon/security/web-security/SUMMARY.md

If a skill file doesn't exist (not installed in this project), skip it and continue. List loaded experts in EXPERTS_LOADED. In EXPERT_DECISIONS, show each specific test decision an expert drove.

Step 1: Find Code

1. Find Code - Identify what needs testing
2. Write Tests - Create test files with real assertions
3. Mock Audit - Before running, review every mock/stub/spy in your tests:
   • What module does this mock replace?
   • Is that module the thing being tested or a dependency of the thing being tested?
   • If you mocked the thing being tested, delete the mock and test the real implementation
   • Example: testing keychain.ts → mock crypto.ts (dependency) ✓, mock storage.ts I/O (dependency) ✓, mock keychain.ts itself ✗
   • For I/O-heavy code: use temp directories and real filesystem operations where possible. Integration tests that touch real I/O catch bugs that mocked tests hide.
4. Run Tests - Execute and verify they pass
5. Report - Document what was tested

REQUIRED Output Format

## Tests: [target]

TESTS_WRITTEN:
- src/__tests__/user.test.ts: [test descriptions]
- src/__tests__/auth.test.ts: [test descriptions]

TESTS_RUN: yes (MANDATORY)
TESTS_PASSED: N
TESTS_FAILED: 0 (must be zero)
TEST_COUNT: N

MOCK_AUDIT:
- [module under test]: mocks [dependency] ✓ | mocks self ✗ (FIXED)
- [module under test]: no mocks, tests real implementation ✓

COVERAGE:
- createUser: tested (happy path, validation error, duplicate email)
- validateToken: tested (valid, expired, malformed)

EXPERTS_LOADED: [list of skill names actually read]
EXPERT_DECISIONS:
- [expert-skill]: [specific test decision it drove]

TESTING_COMPLETE

Validation (Phase will FAIL if violated)

• TEST_COUNT: 0 (no tests written)
• TESTS_RUN: no (tests not executed)
• TESTS_FAILED > 0 (failing tests)

🛑 MANDATORY STOP

After testing:

• DO NOT proceed to next phase
• DO NOT continue with "let me also..."

Your turn ends here. Output TESTING_COMPLETE and STOP.