qa-test-reviewer

name	qa-test-reviewer
description	Code review for test files analyzing structure, assertions, coverage, anti-patterns, and providing a code-smell catalog with before/after improvement examples.
output_dir	reports/reviews

QA Test Reviewer

Purpose

Perform code review on test files to assess quality, identify anti-patterns, and recommend improvements. Analyzes test structure, assertion quality, coverage alignment, naming conventions, and mocking correctness. Produces a structured review report with severity ratings and suggested fixes.

Trigger Phrases

• "Review my test code" / "Code review for tests"
• "Analyze test quality" / "Test code quality review"
• "Find test anti-patterns" / "Test code smells"
• "Review [test file] for best practices"
• "Improve test assertions" / "Fix test structure"
• "Test coverage alignment check"
• "DRY vs DAMP in tests"

Review Criteria

Criterion	What to Check
Test structure	describe/it nesting, setup/teardown placement, logical grouping
Assertion quality	Specific vs vague assertions; one logical assertion per test
Coverage alignment	Test ↔ requirement traceability; tests map to intended behavior
Anti-patterns	Hardcoded data, shared state, sleep waits, implementation coupling
Naming conventions	Descriptive test names; consistent style (should/expect format)
DRY vs DAMP	Shared setup where appropriate; readable tests over excessive abstraction
Mocking correctness	Mocks verify behavior; no over-mocking; correct setup/teardown

See references/anti-patterns.md for the full catalog with before/after examples.

Code Smell Catalog

Smell	Description	See
Magic numbers in tests	Unexplained literals (timeouts, IDs, counts)	references/anti-patterns.md
Copy-paste test duplication	Nearly identical tests; should use parametrize/fixtures	references/anti-patterns.md
Assertion-free tests	Tests that run code but never assert	references/anti-patterns.md
Over-mocking	Mocking everything; tests verify mocks, not behavior	references/anti-patterns.md
Test logic (if/for in tests)	Conditional logic inside tests; tests should be deterministic	references/anti-patterns.md
Ignored tests without explanation	skip/xfail/todo with no reason	references/anti-patterns.md
Hardcoded URLs/selectors	Brittle strings; should use constants or config	references/anti-patterns.md

Each smell includes before/after examples in references/anti-patterns.md.

Output Format

Review Report Template

# Test Review Report: [file/module name]

## Summary
- **Files reviewed:** [count]
- **Issues found:** [count] (Critical: X, High: Y, Medium: Z, Low: W)
- **Overall assessment:** Pass / Needs improvement / Fail

## Issues

### [Severity] [Issue title]
**Location:** [file:line]
**Smell:** [from catalog]
**Description:** [what's wrong]
**Suggested fix:**
[before/after code or instructions]

## Recommendations
1. [Priority action]
2. [Secondary action]

## References
- Anti-patterns: references/anti-patterns.md
- Checklist: references/review-checklist.md

Severity Levels

Level	When to use
Critical	Tests are broken, misleading, or will cause false confidence
High	Significant maintainability or reliability issues
Medium	Best-practice violations; improves quality
Low	Style or minor improvements

Workflow

1. Receive test files — User provides path(s) or content
2. Apply review checklist — Use references/review-checklist.md systematically
3. Identify smells — Map findings to references/anti-patterns.md catalog
4. Assign severity — Critical/High/Medium/Low per impact
5. Suggest fixes — Provide before/after examples where applicable
6. Produce report — Structured output per template above

Scope

Can do (autonomous):

• Review test files for structure, assertions, anti-patterns
• Produce review report with severity and suggested fixes
• Reference anti-patterns catalog and review checklist
• Suggest refactors with before/after examples

Cannot do (requires confirmation):

• Apply fixes directly to test code (review only; user applies)
• Change test framework or migrate tests
• Override organizational coding standards

Will not do (out of scope):

• Modify production/application code
• Execute tests or generate coverage (consume existing reports)
• Implement fixes; this skill reviews and recommends only

References

Topic	File
Anti-patterns catalog with before/after examples	references/anti-patterns.md
Comprehensive test review checklist	references/review-checklist.md

Quality Checklist

• All review criteria from checklist applied
• Issues mapped to code smell catalog where applicable
• Severity assigned per impact (Critical/High/Medium/Low)
• Suggested fixes include before/after when helpful
• Report follows template format
• No hardcoded secrets in examples
• Traceability preserved (test case IDs, requirement links)

Troubleshooting

Symptom	Likely Cause	Fix
Report too verbose	Every minor issue listed	Focus on High/Critical; group Low by category
Before/after unclear	Example too abstract	Use concrete code from reviewed file
Checklist incomplete	Missing framework-specific checks	See review-checklist.md for Jest/pytest/Playwright sections
Severity inconsistent	Subjective judgment	Use severity table; Critical = broken/misleading
User wants fixes applied	Skill is review-only	Clarify scope; offer to generate patch as separate step