dev-devops

Define a workflow with test, build, and deploy jobs: run tests first, build and push a Docker image tagged with the commit SHA, scan it with Trivy, then deploy to staging and production using environment protection rules. The Skill generates this full pipeline configuration.

Use a multi-stage build: install dependencies and build in a builder stage, then copy only artifacts into a slim runtime image. Add a non-root USER, a HEALTHCHECK instruction, pinned base image versions, and a .dockerignore excluding node_modules and .git.

Blue-green runs two full environments and switches traffic instantly for fast rollback, while canary shifts traffic gradually (e.g., 10% then 50%) to limit blast radius. Rolling updates replace pods incrementally and suit routine low-risk changes.

Yes. It covers GitLab CI patterns including workflow rules, needs-based DAGs, caching versus artifacts, OIDC cloud authentication, merge request pipelines, and reusable versioned CI/CD components, alongside GitHub Actions and Jenkins.

No. The Skill explicitly requires presenting a deployment summary and rollback plan, then obtaining explicit user approval before any production or customer-facing deployment. Without approval it stops with a blocked status.

Run kubectl rollout undo deployment/<name> to return to the previous revision, then verify with kubectl rollout status and health checks. For GitOps setups, use argocd app rollback, and for Terraform, revert the configuration and re-apply.