| name | qa-risk |
| description | [QA Process · ISO 31000 + ISTQB CTFL v4.0] Risk-based test prioritization — builds a risk matrix (likelihood × impact), ranks test areas by priority, and flags critical paths for performance testing. |
Helix — QA Risk-Based Prioritization
📚 Knowledge References (loaded automatically):
qa-risk-patterns.md — ISO 31000 risk framework, ISTQB risk-based testing, likelihood/impact matrices
Risk-based testing ensures the highest-risk areas get the most test coverage. Fail fast on what matters most.
Step 1: Identify Risk Areas
Scan the codebase and ask:
1. What are the most business-critical features?
2. What changed most in this release?
3. What broke in previous releases?
4. What has the lowest test coverage now?
5. Any third-party integrations or external dependencies?
Step 2: Build Risk Matrix
Score each area: Likelihood (1–4) × Impact (1–4) = Risk Score
| Feature / Area | Likelihood | Impact | Score | Priority |
|-----------------------|------------|--------|-------|----------|
| Payment processing | 2 | 4 | 8 | 🔴 High |
| User authentication | 2 | 4 | 8 | 🔴 High |
| Search / filtering | 3 | 2 | 6 | 🟡 Med |
| Admin dashboard | 1 | 3 | 3 | 🟢 Low |
Priority mapping:
- Score 8–16 → 🔴 High — test every release, include in Tier 3 performance if critical
- Score 4–7 → 🟡 Medium — test on change
- Score 1–3 → 🟢 Low — test on major releases only
Step 3: Map to Test Tiers
For each High-priority area, assign mandatory test coverage:
| Area | Unit | Integration | E2E | Security | Load Test |
|--------------------|------|-------------|-----|----------|-----------|
| Payment processing | ✅ | ✅ | ✅ | ✅ | ✅ flag |
| Authentication | ✅ | ✅ | ✅ | ✅ | — |
Flag for Tier 3 Performance: Areas with Score ≥ 8 AND high concurrent usage → add to /helix test-perf-load scope.
Step 4: Output
Save to docs/qa/qa-risk-YYYY-MM-DD.md:
- Risk matrix table
- Priority list (High → Medium → Low)
- Tier 3 performance flags
- Recommended test execution order
Done
Present risk matrix + flag list.
If any Tier 3 flags exist, remind user: run /helix test-perf-load separately with flagged endpoints.
Suggest next step: proceed with test execution tiers or /helix qa-explore
Self-Evaluation Loop
1. Risk matrix ครอบคลุม business-critical features ไหม?
2. Score calculation สม่ำเสมอไหม?
3. Tier 3 flags สมเหตุสมผลไหม?
4. มี area ที่ลืมไปไหม?