AZANIR/qa-coverage-analyzer

QA Coverage Analyzer

Purpose

Perform multi-dimensional test coverage analysis across requirements traceability, test design techniques, and code coverage. Identify gaps, prioritize by risk, and produce actionable recommendations for improving test effectiveness.

Trigger Phrases

• "Analyze test coverage" / "Coverage analysis for [project/module]"
• "Requirements traceability matrix" / "RTM coverage"
• "Technique coverage" / "ISO 29119-4 coverage"
• "Code coverage report" / "Line/branch coverage analysis"
• "Coverage gaps" / "Uncovered requirements"
• "Coverage dashboard" / "Coverage heatmap"
• "Risk-based coverage recommendations"

Three Coverage Dimensions

| Dimension | What It Measures | Key Artifacts | |-----------|------------------|----------------| | 1. Requirements/Model | Which requirements have tests; RTM completeness | RTM (Req → Model → Test → Execution) | | 2. Technique | Which ISO 29119-4 techniques applied | EP, BVA, decision tables, state transitions | | 3. Code | Line, branch, condition, function coverage | Istanbul, V8, JaCoCo, coverage.py, SonarQube |

See references/coverage-dimensions.md for detailed measurement methods.

Dimension 1: Requirements/Model Coverage

Model: Requirement → Test Model → Test Case → Execution

• Track: Which requirements have linked test models and test cases
• Identify gaps: Requirements with no tests, tests with no execution, orphan tests
• Metrics: % requirements covered, % executed, traceability completeness

Inputs: Requirements doc, RTM, test case inventory, execution results (Zephyr, TestRail, etc.)

Dimension 2: Technique Coverage (ISO 29119-4)

Techniques: Equivalence Partitioning (EP), Boundary Value Analysis (BVA), Decision Tables, State Transition, Use Case, Classification Trees

• Track: Which techniques applied per requirement/module
• Identify gaps: Requirements tested with only one technique; modules with technique imbalance
• Metrics: Technique distribution, technique-per-requirement ratio

Inputs: Test case metadata (technique tags), requirements, test design docs

Dimension 3: Code Coverage

Metrics: Line, branch, condition, function coverage

| Language/Stack | Tool | Output Format | |----------------|------|---------------| | JavaScript/TypeScript | Istanbul (c8, nyc), V8 | lcov, json-summary | | Java | JaCoCo | xml, html | | Python | coverage.py | xml, html, json | | SonarQube | SonarQube | Unified dashboard |

See references/tools.md for setup and integration.

Outputs

1. Coverage Dashboard — Summary of all three dimensions with percentages and trends
2. Coverage Gaps List — Uncovered requirements, low-technique areas, low-code-coverage modules
3. Risk-Based Recommendations — High-risk uncovered areas prioritized; suggested actions
4. Heatmap by Module — Visual (via qa-diagram-generator) showing coverage intensity per module

Dashboard Template

# Coverage Dashboard

## Summary
| Dimension | Coverage | Status |
|-----------|----------|--------|
| Requirements | X% | ✅/⚠️/❌ |
| Technique | X techniques avg | ✅/⚠️/❌ |
| Code | Line X%, Branch Y% | ✅/⚠️/❌ |

## Gaps
- [Req ID] Uncovered
- [Module] Low technique diversity
- [File/Module] Low code coverage

## Risk-Based Recommendations
1. [High] ...
2. [Medium] ...
3. [Low] ...

Workflow

1. Gather inputs: RTM, test cases, execution results, code coverage reports
2. Analyze each dimension: Compute coverage per dimension (see references/coverage-dimensions.md)
3. Identify gaps: Uncovered requirements, technique-poor areas, low-code-coverage modules
4. Prioritize by risk: Use risk matrix (see qa-test-strategy references/risk-matrix.md); high-risk + low coverage = top priority
5. Generate outputs: Dashboard, gaps list, recommendations, heatmap
6. Historical tracking: Use Memory MCP to store trends for comparison over time

MCP Tools Used

• Sequential Thinking MCP: For decomposition of complex coverage analysis; reconciling conflicting signals across dimensions; prioritizing gaps.
• Memory MCP: For historical coverage tracking; trend comparison (sprint-over-sprint, release-over-release).

Integration with Other Skills

| Need | Skill | Usage | |------|-------|-------| | RTM, test cases from requirements | qa-testcase-from-docs | Source for requirements/model coverage | | Test design techniques | qa-testcase-from-docs | references/test-design-techniques.md | | Risk matrix | qa-test-strategy | references/risk-matrix.md | | Heatmap, charts | qa-diagram-generator | Coverage heatmap, quadrant charts | | NFR thresholds | qa-nfr-analyst | Coverage targets as NFRs |

Scope

Can do (autonomous):

• Analyze coverage across all three dimensions from provided artifacts
• Produce coverage dashboard, gaps list, risk-based recommendations
• Call qa-diagram-generator for heatmaps and charts
• Use Sequential Thinking for complex analysis; Memory for historical tracking
• Reference coverage-dimensions, tools, best-practices

Cannot do (requires confirmation):

• Change coverage thresholds set by stakeholders
• Override organizational coverage policy
• Modify test code or production code

Will not do (out of scope):

• Execute tests or generate coverage reports (consume existing reports)
• Implement fixes for coverage gaps
• Deploy or modify production systems

Quality Checklist

• [ ] All three dimensions analyzed (requirements, technique, code)
• [ ] Gaps list includes requirement IDs, modules, and file paths where applicable
• [ ] Recommendations prioritized by risk (high/medium/low)
• [ ] Dashboard includes percentages and status indicators
• [ ] Heatmap generated for module-level view when requested
• [ ] References to coverage-dimensions, tools, best-practices correct
• [ ] No hardcoded secrets

Troubleshooting

| Symptom | Likely Cause | Fix | |---------|--------------|-----| | Missing RTM | No traceability in place | Run qa-testcase-from-docs to build RTM; ask for requirement IDs | | Technique tags absent | Test cases not tagged | Add technique metadata; use test-design-techniques reference | | Code coverage format unknown | Tool output not recognized | Check tools.md for supported formats; request lcov/xml | | All dimensions green but quality issues | Coverage ≠ quality | Emphasize technique diversity; recommend exploratory testing | | Recommendations too generic | Insufficient risk context | Request risk matrix; use risk-matrix reference | | Historical comparison fails | No prior data in Memory | Store current run in Memory for future comparison |

Reference Files

| Topic | Reference | |-------|-----------| | Coverage dimensions | references/coverage-dimensions.md | | Coverage tools | references/tools.md | | Best practices | references/best-practices.md |