---

name: router-operations description: Master orchestration for routing QA, testing, DevOps, observability, and git workflow questions through 18 operational skills metadata: version: "1.3"

Router: Operations

Master orchestrator that routes quality assurance, testing, deployment, observability, and git workflow questions through the complete operations skill set.

---

Decision Tree: Where to Start?

CopyOPERATIONS QUESTION
    │
    ├─► "How to test this?" ────────────► qa-testing-strategy
    │                                      └─► Unit, integration, E2E, BDD
    │
    ├─► "Write Playwright tests" ───────► qa-testing-playwright
    │                                      └─► E2E, page objects, CI/CD
    │
    ├─► "Test iOS app" ─────────────────► qa-testing-ios
    │                                      └─► XCTest, simulator, UI testing
    │
    ├─► "Test mobile app" ──────────────► qa-testing-mobile
    │                                      └─► iOS + Android, device matrix
    │
    ├─► "Test Android app" ────────────► qa-testing-android
    │                                      └─► Espresso, UIAutomator, Compose
    │
    ├─► "API contract testing" ─────────► qa-api-testing-contracts
    │                                      └─► OpenAPI, GraphQL, gRPC
    │
    ├─► "Debug this issue" ─────────────► qa-debugging
    │                                      └─► Troubleshooting, logging, profiling
    │
    ├─► "Improve error handling" ───────► qa-resilience
    │                                      └─► Circuit breakers, retries, chaos
    │
    ├─► "Setup monitoring" ─────────────► qa-observability
    │                                      └─► OpenTelemetry, tracing, SLOs
    │
    ├─► "Refactor this code" ───────────► qa-refactoring
    │                                      └─► Code smells, tech debt, patterns
    │
    ├─► "Check docs coverage" ──────────► qa-docs-coverage
    │                                      └─► Gap analysis, coverage reports
    │
    ├─► "Test LLM agent" ───────────────► qa-agent-testing
    │                                      └─► Test suites, scoring, refusals
    │
    ├─► "Deploy to production" ─────────► ops-devops-platform
    │                                      └─► K8s, Terraform, GitOps, SRE
    │
    ├─► "Write commit message" ─────────► git-commit-message
    │                                      └─► Conventional commits
    │
    ├─► "Git workflow / branching" ─────► git-workflow
    │                                      └─► GitHub Flow, PRs, reviews
    │
    ├─► "Create documentation" ─────────► docs-codebase
    │                                      └─► README, API docs, ADRs
    │
    ├─► "Write PRD for AI agent" ───────► docs-ai-prd
    │                                      └─► CLAUDE.md, agentic planning
    │
    └─► "Full operations audit" ────────► COMPREHENSIVE ANALYSIS
                                           └─► All operational dimensions

---

Domain Detection

Domain 1: TESTING & QA

Triggers: "test", "testing", "QA", "quality", "coverage", "assertion", "mock", "E2E", "unit test", "integration test", "API test", "contract test"

Primary Skills:

Skill Chain - Complete Test Setup:

Copyqa-testing-strategy (strategy) → qa-testing-playwright (E2E)
    → qa-testing-ios (iOS) → qa-testing-android (Android)
    → qa-testing-mobile (cross platform) → qa-api-testing-contracts (API)
    → qa-agent-testing (AI agents)

Domain 2: DEBUGGING & RESILIENCE

Triggers: "debug", "error", "bug", "crash", "exception", "retry", "circuit breaker", "timeout", "failure"

Primary Skills:

Skill Chain - Production Issues:

Copyqa-debugging (identify issue) → qa-resilience (prevent recurrence)
    → qa-refactoring (improve code quality)

Domain 3: OBSERVABILITY & MONITORING

Triggers: "monitor", "metrics", "tracing", "logging", "SLO", "SLI", "alerting", "APM", "OpenTelemetry"

Primary Skills:

Skill Chain - Observability Setup:

Copyqa-observability (instrumentation) → ops-devops-platform (infrastructure)
    → qa-debugging (incident response)

Domain 4: DEPLOYMENT & INFRASTRUCTURE

Triggers: "deploy", "Kubernetes", "Docker", "Terraform", "CI/CD", "GitOps", "infrastructure", "SRE"

Primary Skills:

Skill Chain - Production Deployment:

Copygit-workflow (branching strategy) → ops-devops-platform (CI/CD + K8s)
    → qa-observability (monitoring) → qa-resilience (failure handling)

Domain 5: GIT & VERSION CONTROL

Triggers: "git", "commit", "branch", "PR", "pull request", "merge", "rebase", "code review"

Primary Skills:

Skill Chain - Git Workflow:

Copygit-workflow (strategy) → git-commit-message (commits)
    → software-code-review (PR reviews)

Domain 6: DOCUMENTATION

Triggers: "documentation", "README", "API docs", "ADR", "PRD", "changelog", "docs", "CLAUDE.md", "AGENTS.md", "large codebase", "project memory"

Primary Skills:

Skill Chain - Documentation:

Copyqa-docs-coverage (audit) → docs-codebase (write docs)
    → docs-ai-prd (if AI project)
    → claude-code-project-memory (if large codebase)

Skill Chain - Large Codebase Setup (100K-1M LOC):

Copyclaude-code-project-memory (hierarchical CLAUDE.md)
    → qa-docs-coverage (documentation audit)
    → docs-codebase (fill gaps)

---

Skill Registry

Testing & QA (8)

Debugging & Quality (3)

Observability (1)

DevOps & Infrastructure (1)

Git & Version Control (2)

Documentation (4)

---

Routing Logic

Keyword-Based Routing

CopyKEYWORDS -> SKILL MAPPING

"test strategy", "test pyramid", "coverage" -> qa-testing-strategy
"Playwright", "E2E", "browser test", "page object" -> qa-testing-playwright
"iOS test", "XCTest", "simulator", "xcrun" -> qa-testing-ios
"Android test", "Espresso", "UIAutomator", "Compose test" -> qa-testing-android
"mobile test", "cross-platform test", "device matrix", "Appium" -> qa-testing-mobile
"API test", "contract test", "OpenAPI", "schema validation" -> qa-api-testing-contracts
"Pact", "consumer driven contract", "CDC" -> qa-api-testing-contracts
"agent test", "LLM test", "persona test", "refusal" -> qa-agent-testing
"docs coverage", "undocumented", "gap analysis" -> qa-docs-coverage
"QAOps", "quality ops", "predictive quality" -> qa-testing-strategy

"debug", "troubleshoot", "stack trace", "profiling" -> qa-debugging
"resilience", "circuit breaker", "retry", "chaos" -> qa-resilience
"refactor", "code smell", "tech debt", "clean code" -> qa-refactoring

"observability", "tracing", "metrics", "SLO", "SLI" -> qa-observability
"OpenTelemetry", "APM", "distributed tracing" -> qa-observability
"DORA metrics", "deployment frequency", "lead time" -> qa-observability
"change failure rate", "MTTR", "mean time to recovery" -> qa-observability
"AIOps", "AI operations", "anomaly detection" -> qa-observability

"deploy", "Kubernetes", "K8s", "Terraform" -> ops-devops-platform
"Docker", "container", "GitOps", "ArgoCD" -> ops-devops-platform
"CI/CD", "pipeline", "GitHub Actions" -> ops-devops-platform
"SRE", "reliability", "incident" -> ops-devops-platform
"platform engineering", "IDP", "developer platform" -> ops-devops-platform
"FinOps", "cloud cost", "cost optimization" -> ops-devops-platform

"commit message", "conventional commit" -> git-commit-message
"git workflow", "branching", "PR", "code review" -> git-workflow
"merge", "rebase", "trunk-based" -> git-workflow

"README", "API docs", "ADR", "changelog" -> docs-codebase
"PRD", "AI agent spec" -> docs-ai-prd
"CLAUDE.md", "AGENTS.md", "large codebase", "project memory", "100K LOC", "documentation setup" -> claude-code-project-memory

Context-Based Routing

---

Skill Chain Patterns

Pattern 1: Complete Test Strategy

CopySTART
  │
  ▼
qa-testing-strategy ────────► Test pyramid + frameworks
  │
  ├─────────────────────────────────────┐
  ▼                                     ▼
qa-testing-playwright ► E2E    qa-testing-ios ► Mobile
  │                                     │
  └─────────────────┬───────────────────┘
                    ▼
           qa-agent-testing ──► AI agents (if applicable)
                    │
                    ▼
              COMPLETE TEST SUITE

Pattern 2: Production Readiness

CopySTART
  │
  ▼
qa-testing-strategy ────────► Tests pass
  │
  ▼
qa-resilience ──────────────► Error handling
  │
  ▼
qa-observability ───────────► Monitoring setup
  │
  ▼
ops-devops-platform ────────► CI/CD + deployment
  │
  ▼
git-workflow ───────────────► Release process
  │
  ▼
PRODUCTION READY

Pattern 3: Incident Response

CopyINCIDENT
  │
  ▼
qa-debugging ───────────────► Root cause analysis
  │
  ▼
qa-observability ───────────► Check metrics/traces
  │
  ▼
qa-resilience ──────────────► Prevent recurrence
  │
  ▼
qa-refactoring ─────────────► Fix underlying issues
  │
  ▼
docs-codebase ──────────────► Post-mortem doc
  │
  ▼
INCIDENT RESOLVED

Pattern 4: LLM Agent Testing

CopySTART
  │
  ▼
qa-agent-testing ───────────► Test suite design
  │
  ├─► Must-ace tasks ────────► Core functionality
  ├─► Refusal edge cases ────► Safety boundaries
  ├─► Scoring rubric ────────► 6-dimension evaluation
  │
  ▼
ai-prompt-engineering ──────► Prompt improvements
  │
  ▼
VALIDATED AGENT

Pattern 5: Documentation Audit

CopySTART
  │
  ▼
qa-docs-coverage ───────────► Gap analysis
  │
  ▼
docs-codebase ──────────────► Write missing docs
  │
  ▼
docs-ai-prd ────────────────► AI agent specs (if applicable)
  │
  ▼
git-workflow ───────────────► PR with docs
  │
  ▼
DOCS COMPLETE

Pattern 6: Large Codebase Documentation (100K-1M LOC)

CopySTART
  │
  ▼
claude-code-project-memory ─► Hierarchical CLAUDE.md structure
  │                           (root + subdirectory docs)
  ├─► Root CLAUDE.md ─────────► Architecture, conventions, entry points
  ├─► Subdirectory docs ──────► Module-specific context (loaded on-demand)
  ├─► AGENTS.md symlink ──────► Cross-platform compatibility
  │
  ▼
qa-docs-coverage ───────────► Audit existing documentation
  │
  ▼
docs-codebase ──────────────► Fill critical gaps
  │
  ▼
LARGE CODEBASE READY FOR LLM ASSISTANCE

Key Guidelines (Jan 2026):

• Keep root CLAUDE.md under ~300 lines
• Use hierarchical structure: subdirectory docs auto-load when needed
• Symlink strategy: ln -s AGENTS.md CLAUDE.md for cross-platform
• Official sources: Anthropic Claude Code Best Practices, OpenAI AGENTS.md Guide

---

Comprehensive Analysis Mode

For full operations audit, invoke skills in parallel:

Layer 1: Quality Assessment (Parallel)

Layer 2: Operational Readiness

Layer 3: Process

---

Cross-Router Handoffs

To router-engineering

When user needs implementation:

• "Build the fix" → Route to software-backend or software-frontend
• "Implement retry logic" → Route to software-backend
• "Add UI for monitoring" → Route to software-frontend

To router-startup

When user shifts to business:

• "Is this ready to launch?" → Route to startup-go-to-market
• "How to price monitoring service?" → Route to startup-business-models

From router-engineering

When router-engineering detects ops needs:

• "How do I test this?" → Route here
• "Deploy to production" → Route here
• "Setup CI/CD" → Route here

---

Quality Gates

Pre-Deployment Checklist

Post-Deployment Checklist

---

Output Templates

Quick Analysis Output

Copy## Operations Analysis: {{TOPIC}}

**Domain Detected**: {{DOMAIN}}
**Primary Skill**: {{SKILL}}
**Supporting Skills**: {{LIST}}

### Current State
- Tests: {{TEST_STATUS}}
- Monitoring: {{MONITORING_STATUS}}
- Docs: {{DOCS_STATUS}}

### Recommended Actions
1. {{ACTION_1}} - Use {{SKILL}}
2. {{ACTION_2}} - Use {{SKILL}}

### Skills to Invoke
- {{SKILL_1}}: {{WHY}}
- {{SKILL_2}}: {{WHY}}

Operations Audit Output

Copy## Operations Audit: {{PROJECT}}

### Testing
- Strategy: {{STATUS}} - {{NOTES}}
- E2E Coverage: {{%}}
- Agent Tests: {{STATUS}}

### Resilience
- Error Handling: {{STATUS}}
- Circuit Breakers: {{STATUS}}
- Retry Policies: {{STATUS}}

### Observability
- Metrics: {{STATUS}}
- Tracing: {{STATUS}}
- Logging: {{STATUS}}
- SLOs: {{STATUS}}

### Infrastructure
- CI/CD: {{STATUS}}
- Deployment: {{STATUS}}
- Rollback: {{STATUS}}

### Documentation
- README: {{STATUS}}
- API Docs: {{STATUS}}
- ADRs: {{STATUS}}

### Priority Actions
1. {{HIGH_PRIORITY_1}}
2. {{HIGH_PRIORITY_2}}
3. {{HIGH_PRIORITY_3}}

---

Related Skills

Note: For deep QA-specific questions (test strategy, debugging workflows, resilience patterns), consider routing through router-qa for more specialized decision trees and workflow patterns.