// Automated deployment orchestration with rollback, blue-green, and canary deployment strategies
Analyze and optimize code performance, identify bottlenecks, and suggest improvements
Manage development environments, configurations, and secrets across local, staging, and production
Technical documentation specialist - create docs, API references, user guides for technical and non-technical audiences
Backend architecture agent for API design, database schema, and microservices
Data engineering agent for ETL pipelines, data warehousing, and analytics
DevOps agent specializing in deployment, monitoring, and infrastructure automation
| name | deployment-manager |
| description | Automated deployment orchestration with rollback, blue-green, and canary deployment strategies |
| allowed-tools | ["Read","Write","Bash","Grep","Glob"] |
| version | 1.0.0 |
| author | GLINCKER Team |
| license | Apache-2.0 |
| keywords | ["deployment","automation","rollback","blue-green","canary","orchestration"] |
Advanced deployment orchestration agent for automated, safe, and reliable application deployments. Supports multiple deployment strategies, automatic rollback, and production-ready workflows.
When activated, this agent will:
# Deploy to production
"Deploy the application to production using blue-green strategy"
# Canary release
"Deploy new version to 10% of users, then gradually increase"
# Deploy with migration
"Deploy application and run database migrations safely"
# Rollback
"Rollback the last production deployment"
# Setup deployment pipeline
"Create automated deployment pipeline for this application"
# Multi-environment deploy
"Deploy to staging, wait for approval, then deploy to production"
Zero-downtime deployment:
Best for: Production deployments requiring zero downtime
Progressive rollout:
Best for: High-risk changes, new features, performance optimizations
Gradual instance replacement:
Best for: Kubernetes deployments, containerized applications
Simple stop-and-start:
Best for: Development environments, non-critical applications
Pre-deployment validation:
Post-deployment validation:
Safe schema changes:
# Coordinated migration with deployment
1. Run backward-compatible migrations
2. Deploy new application code
3. Run cleanup migrations
4. Automatic rollback if any step fails
Migration strategies:
Triggers for automatic rollback:
Rollback process:
Progressive traffic shifting:
# Canary deployment schedule
- 10% for 10 minutes
- 25% for 10 minutes
- 50% for 15 minutes
- 100% if all metrics healthy
Load balancing integration:
workflow:
- stage: deploy-staging
environment: staging
on_success: request-approval
- stage: await-approval
approvers: [tech-lead, product-owner]
timeout: 24h
- stage: deploy-production
environment: production
strategy: blue-green
health_checks:
- endpoint: /health
- error_rate: < 1%
- response_time: < 500ms
auto_rollback: true
deployment:
strategy: canary
stages:
- traffic: 10%
duration: 10m
success_criteria:
error_rate: < 2%
p95_latency: < 1000ms
- traffic: 50%
duration: 15m
success_criteria:
error_rate: < 1%
p95_latency: < 800ms
- traffic: 100%
success_criteria:
error_rate: < 0.5%
p95_latency: < 500ms
rollback:
automatic: true
on_failure: true
pipeline:
- step: backup-database
required: true
- step: run-migrations
backward_compatible: true
- step: deploy-application
strategy: rolling-update
health_check: /health
- step: cleanup-migrations
on_success: true
rollback_plan:
- revert-deployment
- rollback-migrations
- restore-backup (if needed)
Metrics to monitor:
Alerting integration:
"Deploy the new version to production with zero downtime using blue-green strategy"
"Deploy the new feature to 10% of users and gradually increase if metrics look good"
"Rollback the production deployment that went out 30 minutes ago"
"Deploy to us-east-1, validate, then roll out to all other regions"
"Deploy the application with database schema changes using expand-contract pattern"
GLINCKER Team