container-deployment_skill

Overview

This skill designs and implements containerization and deployment automation for Brainarr, enabling easy distribution, scaling, and reliable upgrades. It focuses on building optimized Docker images, publishing multi-architecture artifacts to registries, and automating deployments with CI/CD, Kubernetes, and Helm. The goal is repeatable, minimal-effort installs for both pre-packaged and plugin-only deployment models.

How this skill works

I create efficient Dockerfiles (pre-packaged Lidarr+Brainarr and plugin-only images), configure multi-arch builds, and wire publishing to registries like GHCR. I add CI workflows to build, tag, and push images, plus Helm charts and Kubernetes manifests for orchestration. Deployment strategies (blue-green, canary, rolling) and IaC examples (Terraform, docker-compose, Ansible) are provided to automate provisioning and environment-specific configuration.

When to use it

• You need a one-command deployable image combining Lidarr and Brainarr
• You want a minimal plugin-only container for mounting into an existing Lidarr instance
• You need multi-architecture images for amd64 and arm platforms
• You want automated image builds and publishing on release via GitHub Actions
• You plan to deploy to Kubernetes and need Helm charts and manifests

Best practices

• Use multi-stage Dockerfiles and layer caching to minimize image size
• Publish multi-arch images (linux/amd64, linux/arm64, linux/arm/v7) with clear tag strategy
• Provide both pre-packaged and plugin-only images to support different deployment models
• Include healthchecks, liveness/readiness probes, and sensible start-periods for Lidarr
• Automate CI to extract version from tags and push to GHCR with cache-to/cache-from for fast rebuilds

Example use cases

• Build a combined Lidarr+Brainarr image and publish ghcr.io/richertunes/brainarr:latest for one-command deploys
• Produce a lightweight plugin-only image to mount into a host Lidarr container or sidecar
• Create a GitHub Actions workflow that builds multi-arch images and tags by semantic version on release
• Deploy Brainarr with Helm to a Kubernetes cluster with PVC-backed config and music volumes and liveness probes
• Implement a blue-green or canary rollout to validate plugin updates before full traffic cutover