菜单
Use when measuring and reporting QA quality — defect escape rate, test coverage analysis, flaky test rate, mean time to detect, shift-left metrics, and building quality dashboards for stakeholders.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | qc-metrics |
| description | Use when measuring and reporting QA quality — defect escape rate, test coverage analysis, flaky test rate, mean time to detect, shift-left metrics, and building quality dashboards for stakeholders. |
Defect Escape Rate (most important)
Formula: Production defects / (Pre-prod defects + Production defects) × 100
Target: < 10% (less than 10% of defects found in production)
Why it matters: Measures testing effectiveness; high rate = gaps in coverage
Test Coverage
Line coverage: % of code lines executed by tests
Branch coverage: % of code branches executed
Requirement coverage: % of requirements with test cases
Target: > 80% branch coverage for business-critical paths
Caution: Coverage is a floor not a ceiling; 100% with bad tests is worthless
Defect Detection Distribution
% defects found per phase: Unit test | Integration | QA | UAT | Production
Target: Most defects found in unit/integration (shift-left)
Cost multiplier: Defect in prod costs ~30x more than unit test phase
Flaky Test Rate
Formula: Tests that failed non-deterministically in last 7 days / total tests
Target: < 2% flaky rate
Impact: High flaky rate erodes trust in test suite, developers start ignoring failures
Mean Time to Detect (MTTD)
Time from bug introduction (code commit) to detection
Target: < 24 hours for critical paths
Enables: Faster feedback, cheaper fixes
Test Execution Time
Time from code commit to test results
Target: < 10 min for unit+integration, < 30 min for full suite
Impact: Slow tests = developers don't run them locally
# Monthly defect escape rate report
def calculate_escape_rate(month: str) -> dict:
pre_prod_defects = get_defects_found_before_production(month)
prod_defects = get_production_incidents(month)
total = len(pre_prod_defects) + len(prod_defects)
escape_rate = len(prod_defects) / total * 100 if total > 0 else 0
return {
"month": month,
"pre_prod_defects": len(pre_prod_defects),
"production_defects": len(prod_defects),
"escape_rate": f"{escape_rate:.1f}%",
"trend": "improving" if escape_rate < get_previous_month_rate() else "degrading"
}
Healthy test pyramid:
E2E tests: 5-10% (slow, expensive, cover critical paths)
Integration tests: 20-30% (medium speed, component interactions)
Unit tests: 60-70% (fast, cheap, cover business logic)
Unhealthy patterns:
Ice cream cone (inverted pyramid):
Many E2E, few unit tests → slow CI, brittle suite
Fix: Identify what E2E tests are actually testing unit-level logic
Hourglass (gaps in middle):
Many unit + many E2E, no integration
Fix: Add API/service-level integration tests
Check your pyramid:
grep -r "@pytest.mark" tests/ | grep -c "e2e\|integration\|unit"
Dashboard sections:
1. Build Health (real-time)
- Last build status: PASS/FAIL
- Build duration trend (line chart)
- Flaky test list (top 10)
2. Coverage Trends (weekly)
- Line coverage % over time (trend line)
- Files with lowest coverage (table)
- Coverage delta per PR
3. Defect Metrics (monthly)
- Defect escape rate over time
- Defect distribution by phase (bar chart)
- Open P1/P2 count (KPI card)
- Time to resolution by priority
4. Release Readiness (per release)
- Test completion % by feature
- Outstanding P1/P2 defects
- Test pass rate last 5 runs
- Sign-off status
Measuring shift-left effectiveness:
Before shift-left: 70% bugs found in QA/UAT, 30% in unit tests
After 6 months: 50% bugs in unit, 30% integration, 20% QA, <5% production
Leading indicators (predict future quality):
- PR code review turnaround time (faster = fewer bugs pass through)
- Unit test coverage per PR (requires tests with new code)
- Static analysis violations per PR (code quality proxy)
Lagging indicators (measure past quality):
- Defect escape rate (measures testing effectiveness)
- Production incident rate (measures overall quality)
- Customer-reported bugs (ultimate quality measure)
Report: leading indicators weekly (actionable), lagging monthly (strategic)
qc-defect-management — defect data feeds escape rate and distribution metricsado-pipeline-optimization — test result publishing enables coverage trendingtest-strategy — metrics validate and inform the overall test strategyUse when conducting quality audits — reviewing process compliance, identifying gaps between defined process and actual practice, conducting structured inspections (code review audits, test quality reviews), and producing audit reports with remediation plans.
Use when managing quality risk — identifying quality risks in a product or release, applying risk-based testing prioritization, creating risk mitigation plans, and communicating quality risk to stakeholders for go/no-go decisions.
Use when designing or implementing test automation — choosing the right automation framework (Playwright, pytest, JUnit), Page Object Model, selector strategies, test isolation, managing flaky tests, and CI integration.
Use when managing defects — writing effective bug reports, applying severity/priority matrix, tracking defect lifecycle, conducting root cause analysis, and measuring defect metrics for process improvement.
Use when managing test data — designing test data strategies, using factories and builders, creating fixtures, generating synthetic data, masking PII for testing, and managing test database state.
Use when designing test cases — applying boundary value analysis, equivalence partitioning, decision tables, pairwise testing, and exploratory testing techniques to maximize defect detection with minimal test cases.