How to use the Deployment Patterns agent skill

What does the deployment-patterns skill make the agent do?

The skill gives the agent a set of deployment workflows and best practices. It covers three rollout methods: rolling (the default), blue-green and canary. It also includes multi-stage Dockerfile templates for Node.js, Go and Python/Django apps, a standard GitHub Actions pipeline with test, build and deploy jobs, health check endpoint examples, and checklists of good and bad Docker practices.

The skill is reference material and templates. It does not deploy anything itself. The sample deploy step in the pipeline only prints a message and leaves the real deploy command as a placeholder.

• Compare rolling, blue-green and canary rollouts and their trade-offs
• Write Dockerfiles for Node.js, Go or Python apps
• Set up a CI/CD pipeline with test, build and deploy stages
• Add /health endpoints and Kubernetes probe examples

How do you install the skill?

The skill ships as a ZIP download from the Deployment Patterns page in this directory. The package contains one file, SKILL.md. You can install it in either of two ways.

Option A is a personal install. Unzip the download into ~/.claude/skills/ so the skill is available across your projects. Option B is a project install. Unzip it into the .claude/skills/ folder inside a specific project so it applies only there.

Option C is the Claude apps. Upload the ZIP under Settings → Capabilities → Skills.

• Personal: unzip into ~/.claude/skills/
• Project: unzip into .claude/skills/ in your project folder
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP

How do you trigger the skill with a natural request?

You do not need a special command. Describe a deployment task in plain language and the agent will use the skill when the request matches its scope. The skill is meant for setting up deployment infrastructure, containerizing an app, planning a release and adding health checks.

Be specific about your stack and goal. Naming Node.js, Go or Python and saying whether you want a rollout method or a pipeline gives the agent the context it needs to pick the right template.

• "Write a multi-stage Dockerfile for my Node.js app."
• "Set up a GitHub Actions pipeline that tests, builds and pushes an image on merge to main."
• "Which rollout should I use for a high-traffic checkout service: blue-green or canary?"
• "Add a /health endpoint and a readiness probe to my Express app."

What do the templates look like?

The Node.js Dockerfile uses three stages: install dependencies, build, then a production image that runs as a non-root user and includes a HEALTHCHECK instruction. The GitHub Actions example runs lint, typecheck and tests on every push and pull request, and builds and pushes an image to GHCR only on main.

Treat the templates as starting points. The source pins versions such as node:22-alpine, golang:1.22-alpine and python:3.12-slim, which may be outdated for your project.

What does the skill need and what are the safety notes?

The skill runs no scripts and uses no network access. It only supplies guidance for the agent to read. The examples assume tools you may need to install yourself: Docker for the container examples, GitHub Actions for the CI/CD example, and the Node.js, Go or Python toolchain for the app being deployed.

Review everything before using it in production. Do not push the sample pipeline to a real environment without replacing the placeholder deploy step with your platform's command. Keep secrets out of images and use environment variables or a secrets manager, as the skill's own best practices advise.

• Docker for container examples
• GitHub Actions for the CI/CD example
• Node.js, Go or Python toolchain for your app
• Replace the placeholder deploy step before any real release

What tips help you get the most from it?

Use the skill to plan before you build. Ask about the rollout method that fits your risk level, then ask for the matching Dockerfile and pipeline. Check the output against your own stack, since the templates are examples and not tailored configurations.

Follow the checklist. The skill recommends specific version tags instead of latest, non-root users, a .dockerignore file, and copying dependency files first for layer caching.

• Name your framework and hosting platform in the request
• Ask for the trade-offs of a rollout method before choosing one
• Verify pinned image versions and the deploy command yourself