ci-cd-builder_skill

Overview

This skill helps developers design and implement CI/CD pipelines for GitHub Actions, GitLab CI, Jenkins, and CircleCI. It provides ready-to-use templates, common configuration snippets, and practical deployment patterns for Node.js, Docker, testing, and artifact management. The goal is faster feedback, secure secret handling, and repeatable automated builds and deployments.

How this skill works

The skill inspects project type and recommends pipeline templates and steps tailored to that stack (for example Node.js test/build, Docker build-and-push). It supplies concrete YAML snippets for job orchestration, caching, conditional deployment, and secret injection. It also suggests pipeline structure: parallel test jobs, build artifacts, and dependent deployment jobs, plus notifications and version pinning.

When to use it

• Setting up CI for a new repository (Node.js, Docker, or polyglot projects).
• Adding automated tests, linting, and build stages to existing workflows.
• Configuring Docker image build and push to a registry like GitHub Container Registry.
• Implementing conditional production deploys and environment-specific steps.
• Optimizing pipeline performance with dependency caching and parallel jobs.

Best practices

• Split fast feedback (unit tests, lint) into parallel jobs to reduce latency.
• Cache dependencies (npm, pip, etc.) to speed repeat builds and reduce flakiness.
• Store credentials in platform Secrets and inject them as environment variables at runtime.
• Pin Action and runner tool versions to avoid unexpected breaking changes.
• Publish build artifacts from build jobs and gate deployments on successful tests and security scans.

Example use cases

• Node.js project CI: checkout, setup-node, npm ci, run tests, lint, build, upload artifacts.
• Docker pipeline: buildx build and push images to ghcr.io with credentials from secrets.
• Conditional deploy: run deploy script only on main branch using an if condition.
• Cache dependencies: use actions/cache to reuse npm cache keyed by package-lock.json hash.
• Multi-stage pipeline: tests -> build -> docker push -> deploy, with notifications on failure.