Generate Production-Ready Deployment Guides

This skill provides expertise in creating comprehensive, production-ready deployment guides that enable teams to successfully deploy applications and infrastructure, understanding the critical importance of clear step-by-step instructions, environment-specific configurations, troubleshooting guidance, and operational considerations. Core structure and organization principles cover prerequisites first (system requirements, dependencies, access requirements stated up front), environment separation (distinct instructions for development, staging, and production), step verification (validation steps after each major deployment phase), rollback procedures (documented for every deployment stage), and operational readiness (monitoring, logging, and maintenance procedures). Documentation standards use imperative instructions, exact commands with expected output, explicit tool/dependency/image versions, the reasoning behind each step, and estimated execution time per phase.

The standard deployment guide structure covers an overview (application/architecture description, deployment strategy such as blue-green/rolling/canary, target environments), prerequisites (system requirements, required tools and versions, access requirements, network/security considerations), a pre-deployment checklist, deployment steps organized into infrastructure preparation, application deployment, configuration/testing, and go-live phases, post-deployment verification (health checks, performance validation, security verification), troubleshooting (common issues, log locations, emergency contacts), and rollback procedures (automatic triggers, manual steps, data recovery).

Environment-specific configuration is demonstrated through a Kubernetes production Deployment manifest using RollingUpdate strategy with maxSurge/maxUnavailable settings, resource requests/limits, and liveness/readiness probes, plus a CI/CD deployment script running pre-deployment cluster checks, backing up the current deployment before applying the new one, waiting on rollout status, running health checks in a retry loop, and running smoke tests. Database migration procedures are demonstrated through a SQL migration checklist emphasizing backup-first and staging-first testing, a transactional forward migration adding a column and recording a schema version, with the corresponding rollback kept as a commented reference. Monitoring and alerting setup is demonstrated through a Prometheus alert rules ConfigMap defining a critical alert for the app being down and a warning alert for elevated 5xx error rates.

Best practices cover security considerations (never including secrets in deployment guides, referencing secret management systems instead, documenting security scanning requirements, network/firewall configuration, and required RBAC permissions), performance optimization (resource requirements, scaling parameters, performance testing procedures, monitoring thresholds and SLA requirements, caching configuration), and operational excellence (detailed logging configuration and log analysis guidance, disaster recovery procedures with RTO/RPO requirements, maintenance windows and update procedures, backup/restore procedures with tested protocols). A troubleshooting framework provides a diagnostic toolkit script checking pod/service/ingress status, recent cluster events, application logs, and resource usage, alongside common issue resolutions for image pull errors, health check failures, resource constraints, network connectivity issues, and database connection problems.

Use this skill when writing a production-ready deployment guide - covering prerequisites, environment-specific configuration, CI/CD deployment scripts, database migrations, monitoring setup, and rollback procedures. It is well suited to applications deployed to Kubernetes or similar orchestrated infrastructure needing repeatable, verifiable deployment steps. It is not meant for trivial single-command deployments with no meaningful rollback, verification, or environment-separation needs.

Input: the application's deployment target (Kubernetes, CI/CD pipeline), environments, and rollback/monitoring requirements.

Output: a structured deployment guide with Kubernetes manifests, deployment scripts, migration procedures, and monitoring/alerting configuration. Example Kubernetes rollout status check:

kubectl apply -f k8s/deployment.yaml
kubectl rollout status deployment/webapp-production -n production --timeout=300s

Works with Kubernetes deployment manifests, CI/CD pipelines, SQL-based database migrations, and Prometheus for monitoring and alerting.

DevOps and platform engineers writing production deployment runbooks, and teams that need repeatable, verified deployment procedures with rollback and monitoring built in rather than ad hoc deployment steps.