cicd-pipeline-qe-orchestrator

When to Use

• Designing pipeline test strategies
• Implementing quality gates
• Coordinating multiple QE agents
• Shift-left and shift-right testing

Phase-Agent Matrix

| Phase | Primary Agents | Key Skills |

|-------|---------------|------------|

| Commit | qe-test-generator, qe-requirements-validator | tdd-london-chicago, shift-left |

| Build | qe-test-executor, qe-coverage-analyzer, qe-flaky-test-hunter | test-automation, mutation-testing |

| Test | qe-api-contract-validator, qe-performance-tester, qe-security-scanner | api-testing, performance, security |

| Staging | qe-chaos-engineer, qe-visual-tester, qe-deployment-readiness | chaos-engineering, accessibility |

| Prod | qe-production-intelligence, qe-quality-analyzer | shift-right, compliance |

Quality Gate Thresholds

| Phase | Metric | Threshold | Blocking |

|-------|--------|-----------|----------|

| Commit | Unit coverage | > 80% | Yes |

| Build | All tests pass | 100% | Yes |

| Build | Mutation score | > 70% | No |

| Test | API contracts | No breaking changes | Yes |

| Test | p95 response | < 200ms | Yes |

| Test | Security critical | 0 | Yes |

| Staging | Deployment readiness | > 85% | Yes |

Fleet Configuration

• Topology: hierarchical
• Max Agents: 10 per phase
• Coordination: aqe/* memory namespace

---

Phase 1: Commit (Shift-Left)

Goal: Catch defects early, ensure testability

Agents:

• qe-test-generator - Generate unit tests
• qe-requirements-validator - BDD scenarios, INVEST criteria

Skills: shift-left-testing, tdd-london-chicago, code-review-quality

```javascript

// Parallel execution

Task("Generate Tests", "Create unit tests for new methods", "qe-test-generator")

Task("Validate Requirements", "Check BDD scenarios", "qe-requirements-validator")

```

Gates: Unit coverage > 80%, Static analysis clean, Code review approved

---

Phase 2: Build

Goal: Validate integration, ensure coverage

Agents:

• qe-test-executor - Run test suites
• qe-coverage-analyzer - Coverage gaps (O(log n))
• qe-flaky-test-hunter - Detect/stabilize flaky tests
• qe-regression-risk-analyzer - Minimal regression suite

Skills: test-automation-strategy, mutation-testing, regression-testing

```javascript

Task("Execute Tests", "Run full suite, store in aqe/test-results/*", "qe-test-executor")

Task("Coverage Analysis", "Identify gaps", "qe-coverage-analyzer")

Task("Flaky Detection", "Analyze test history", "qe-flaky-test-hunter")

```

Gates: All tests pass, Coverage > 90% critical paths, No new flaky tests

---

Phase 3: Integration/Test

Goal: Validate contracts, performance, security

Agents:

• qe-api-contract-validator - Breaking changes detection
• qe-performance-tester - Load test critical paths
• qe-security-scanner - SAST/DAST scans
• qe-test-data-architect - Realistic test data (10k+/sec)

Skills: api-testing-patterns, performance-testing, security-testing

```javascript

// Parallel testing

Task("API Contracts", "Validate for breaking changes", "qe-api-contract-validator")

Task("Performance", "Load test 1000 users", "qe-performance-tester")

Task("Security", "SAST/DAST scan", "qe-security-scanner")

```

Gates: No breaking API changes, p95 < 200ms, No critical vulnerabilities

---

Phase 4: Staging

Goal: Validate production-like environment, resilience

Agents:

• qe-chaos-engineer - Fault injection
• qe-visual-tester - Visual regression
• qe-deployment-readiness - Risk assessment

Skills: chaos-engineering-resilience, accessibility-testing, visual-testing

```javascript

Task("Chaos Testing", "Controlled failure injection", "qe-chaos-engineer")

Task("Visual Testing", "Visual regression", "qe-visual-tester")

Task("Deployment Check", "Risk assessment", "qe-deployment-readiness")

```

Gates: Chaos tests pass, No visual regressions, Readiness > 85%

---

Phase 5: Production (Shift-Right)

Goal: Monitor real usage, validate compliance

Agents:

• qe-production-intelligence - Incident → test scenarios
• qe-quality-analyzer - Quality metrics and trends

Skills: shift-right-testing, compliance-testing

```javascript

Task("Production Intelligence", "Convert incidents to tests", "qe-production-intelligence")

Task("Quality Analysis", "Production metrics", "qe-quality-analyzer")

```

Gates: Synthetic monitors pass, Error rate < 0.1%, Compliance validated

---

```javascript

// Phase 1: Commit

Task("TDD Generation", "Generate tests for new features", "qe-test-generator")

Task("Requirements", "Validate BDD scenarios", "qe-requirements-validator")

// Phase 2: Build

Task("Execute Tests", "Full suite with coverage", "qe-test-executor")

Task("Coverage", "Analyze gaps", "qe-coverage-analyzer")

Task("Flaky Hunt", "Stabilize flaky tests", "qe-flaky-test-hunter")

// Phase 3: Integration

Task("API Contracts", "Check breaking changes", "qe-api-contract-validator")

Task("Performance", "1000 user load test", "qe-performance-tester")

Task("Security", "SAST/DAST scans", "qe-security-scanner")

// Phase 4: Staging

Task("Chaos", "Fault injection testing", "qe-chaos-engineer")

Task("Visual", "Visual regression", "qe-visual-tester")

Task("Readiness", "Deployment assessment", "qe-deployment-readiness")

// Phase 5: Production

Task("Intelligence", "Convert incidents", "qe-production-intelligence")

Task("Quality Gate", "Final validation", "qe-quality-gate")

```

---

By Risk Level

| Risk | Strategy | Agents |

|------|----------|--------|

| Critical | All phases, manual gates | Full fleet |

| High | Automated gates, comprehensive | 10+ agents |

| Medium | Smart selection, risk-based | 5-8 agents |

| Low | Minimal regression, fast | 2-3 agents |

By Application Type

| Type | Focus Skills | Primary Agents |

|------|-------------|----------------|

| API | api-testing, contract, performance | api-contract-validator, performance-tester |

| Web UI | visual-testing, accessibility | visual-tester, accessibility |

| Mobile | mobile-testing, compatibility | performance-tester, visual-tester |

| Backend | database-testing, security | security-scanner, performance-tester |

---

Memory Namespace

```

aqe/pipeline/

├── phase-results/* - Results from each phase

├── quality-gates/* - Gate validation results

├── orchestration-plan/* - Selected skills and agents

├── test-plan/generated - Test plans

├── coverage/gaps - Coverage analysis

├── security/findings - Security results

└── performance/results - Performance data

```

Fleet Orchestration

```typescript

Task("Fleet Orchestration",

"Coordinate 10 agents across phases: commit (2), build (3), test (3), staging (2)",

"qe-fleet-commander")

```

Blackboard Events

| Event | Trigger | Subscribers |

|-------|---------|-------------|

| phase:commit:complete | Commit phase done | build agents |

| coverage:gap:detected | Gap found | test-generator |

| security:finding:critical | Vulnerability | quality-gate |

| quality:gate:evaluated | Gate decision | fleet-commander |

---

```json

{

"commit": {

"gates": [

{ "metric": "unit_coverage", "threshold": 80, "blocking": true },

{ "metric": "static_analysis_critical", "max": 0, "blocking": true }

]

},

"build": {

"gates": [

{ "metric": "all_tests_passed", "threshold": 100, "blocking": true },

{ "metric": "mutation_score", "threshold": 70, "blocking": false }

]

},

"integration": {

"gates": [

{ "metric": "api_breaking_changes", "max": 0, "blocking": true },

{ "metric": "performance_p95_ms", "threshold": 200, "blocking": true },

{ "metric": "security_critical", "max": 0, "blocking": true }

]

}

}

```

---

| Issue | Cause | Solution |

|-------|-------|----------|

| OOM during tests | Running all tests in parallel | Use batched execution |

| Pipeline too slow | Comprehensive testing every commit | Smart test selection |

| Gates always failing | Thresholds too strict | Analyze trends, adjust |

---

• [agentic-quality-engineering](../agentic-quality-engineering/) - Fleet coordination
• [shift-left-testing](../shift-left-testing/) - Early defect detection
• [shift-right-testing](../shift-right-testing/) - Production monitoring
• [chaos-engineering-resilience](../chaos-engineering-resilience/) - Resilience testing

---

The CI/CD Pipeline QE Orchestrator provides:

• Phase-based strategy with optimal skill/agent selection
• Quality gates that block bad deployments
• Adaptive strategy based on risk and context
• Full fleet coordination through memory namespaces

With Agents: Use qe-fleet-commander for multi-agent orchestration. Coordinate through aqe/* memory namespace. Batch operations for efficiency.