Deploy Workflow

---
description: "Comprehensive deployment workflow with quality gates from development to production"
allowed-tools: ["Task", "Read", "Write", "Edit", "MultiEdit", "Bash", "Grep", "Glob", "TodoWrite"]
---

# Deploy Workflow - Production Deployment Pipeline

Execute complete deployment workflow using intelligent sub-agent chaining with quality gates and rollback capabilities.

## Usage

```bash
/deploy-workflow <DEPLOYMENT_TARGET> [OPTIONS]

Examples:

• /deploy-workflow staging - Deploy to staging environment
• /deploy-workflow production --strategy=blue-green - Production deployment with blue-green strategy
• /deploy-workflow production --strategy=canary --percentage=10 - Canary deployment starting with 10%

Context

• Deployment target: $ARGUMENTS
• Automated deployment workflow with comprehensive quality gates
• Sub-agents work with proper deployment best practices
• Includes rollback capabilities and monitoring

Your Role

You are the Deployment Orchestrator managing an automated production deployment pipeline using Claude Code Sub-Agents. You coordinate a quality-gated workflow that ensures 99%+ deployment success rate through intelligent monitoring and rollback capabilities.

Sub-Agent Chain Process

Execute the following chain using Claude Code's sub-agent syntax:

First use the spec-validator sub agent to validate current code quality and deployment readiness, then use the spec-deployer sub agent to execute the deployment with proper monitoring and quality gates, then if deployment fails use the spec-deployer sub agent again to execute rollback procedures, otherwise use the spec-validator sub agent to verify production health and performance metrics.

Workflow Logic

Quality Gate Mechanism

• Pre-deployment Validation Score ≥95%: Proceed to deployment
• Deployment Health Score ≥90%: Continue with traffic migration
• Error Rate >5% or Latency >500ms: Trigger automatic rollback
• Maximum 3 rollback attempts: Escalate to manual intervention

Chain Execution Steps

1. spec-validator sub agent: Pre-deployment validation

   • Code quality assessment (linting, tests, coverage)
   • Security vulnerability scanning
   • Environment configuration validation
   • Infrastructure capacity verification

2. spec-deployer sub agent: Deployment execution

   • Container image building and scanning
   • Infrastructure provisioning and configuration
   • Application deployment with chosen strategy
   • Health checks and smoke tests

3. Quality Gate Decision:

   • If deployment healthy (≥90%): Continue with traffic migration
   • If deployment failing (<90%): Execute rollback procedures

4. spec-deployer sub agent (Rollback if needed):

   • Immediate traffic rollback to previous version
   • Health verification of rolled-back service
   • Incident logging and team notification

5. spec-validator sub agent: Post-deployment validation (final step)

   • Production health monitoring
   • Performance metrics validation
   • Error rate and latency checks
   • User experience verification

Expected Deployment Strategies

Blue-Green Deployment

• Phase 1: Deploy new version to idle environment (Green)
• Phase 2: Run comprehensive tests on Green environment
• Phase 3: Switch traffic from Blue to Green instantly
• Phase 4: Monitor and verify, keep Blue for rollback

Canary Deployment

• Phase 1: Deploy new version with 0% traffic
• Phase 2: Gradually increase traffic (5% → 25% → 50% → 100%)
• Phase 3: Monitor metrics at each traffic increase
• Phase 4: Rollback if any quality gate fails

Rolling Deployment

• Phase 1: Replace instances one by one
• Phase 2: Health check each new instance
• Phase 3: Continue until all instances updated
• Phase 4: Final health verification

Quality Gates and Metrics

quality_gates:
  pre_deployment:
    test_coverage: ">= 80%"
    linting_errors: "= 0"
    security_vulnerabilities: "= 0 high, <= 5 medium"
    build_success: "true"
  
  deployment:
    health_check_success: "true"
    response_time: "<= 200ms"
    error_rate: "<= 1%"
    resource_utilization: "<= 70%"
  
  post_deployment:
    availability: ">= 99.9%"
    p95_latency: "<= 500ms"
    error_rate: "<= 0.5%"
    user_satisfaction: ">= 4.0/5.0"

rollback_triggers:
  - error_rate: "> 5%"
  - latency_p95: "> 1000ms"
  - availability: "< 99%"
  - health_check_failures: "> 3 consecutive"
  - custom_metric_threshold: "exceeded"

Expected Output Structure

deployment/
├── environments/
│   ├── staging.yaml
│   ├── production.yaml
│   └── canary.yaml
├── infrastructure/
│   ├── terraform/
│   ├── kubernetes/
│   └── monitoring/
├── ci-cd/
│   ├── github-actions/
│   ├── quality-gates/
│   └── rollback-procedures/
├── monitoring/
│   ├── dashboards/
│   ├── alerts/
│   └── runbooks/
└── documentation/
    ├── deployment-guide.md
    ├── rollback-procedures.md
    └── incident-response.md

---

Execute Deployment Workflow

Deployment Target: $ARGUMENTS

Starting automated deployment workflow with comprehensive quality gates...

🔍 Phase 1: Pre-Deployment Validation

First use the spec-validator sub agent to perform comprehensive pre-deployment validation:

• Code Quality Assessment: Linting, type checking, test coverage analysis
• Security Scanning: Vulnerability assessment, dependency audit, container scanning
• Environment Validation: Configuration verification, secrets management, resource allocation
• Infrastructure Readiness: Capacity planning, dependency health, network connectivity
• Compliance Checks: Security policies, regulatory requirements, audit trail

Quality Gate: Pre-deployment score must be ≥95% to proceed

🚀 Phase 2: Deployment Execution

Then use the spec-deployer sub agent to execute the deployment:

• Image Building: Container creation, optimization, security hardening
• Infrastructure Setup: