@arcaelas/pctl

Declarative pod orchestrator for multi-cloud container deployments.

One YAML. Multiple providers. Zero vendor lock-in.

What is PCTL?

PCTL takes a single configuration file and deploys your containerized services to AWS EKS, Google Cloud GKE, or Docker (local and remote via SSH). It handles building images, pushing to registries, creating Kubernetes manifests, managing state, and cleaning up resources.

Features

• Declarative - Define services, scaling, storage, and health checks in YAML
• Multi-provider - Deploy to AWS, GCP, or Docker from the same config
• Resolvers - Dynamic values with ${env:KEY}, ${ssm:/path}, ${cfn:export}, ${self:path}
• Plugins - Extensible pipeline for custom transformations
• Diff-based - Only redeploys what changed (fingerprinting)
• State tracking - JSON state file for clean destroy
• Registry agnostic - ECR, Artifact Registry, GHCR, Docker Hub
• Auto-scaling - HPA support for Kubernetes providers
• RBAC - ServiceAccount, Role, and RoleBinding management

Quick Start

npm install -g @arcaelas/pctl

Create pctl.yaml:

name: my-app

services:
  api:
    image: ./Dockerfile
    scale:
      replica: 1
      cpu: 256m
      memory: 512Mi
    ports:
      - 3000
    provider:
      name: docker
      options: {}

Deploy and destroy:

pctl deploy
pctl destroy

Providers

| Provider | Target | Auth | Storage | |---|---|---|---| | aws | EKS (Kubernetes) | IAM / STS / env vars | EBS / EFS | | gcp | GKE (Kubernetes) | gcloud CLI | PD / Filestore | | docker | Local or SSH | Docker socket / SSH key | Docker volumes |

Configuration

name: my-stack

resolver: []                    # Custom resolver modules
plugin: []                      # Custom plugin modules
custom: {}                      # Reusable values store

services:
  my-service:
    image: ./Dockerfile         # or nginx:latest
    registry: ghcr.io/org/repo  # or { url, username, password }
    command: "node server.js"   # Override CMD
    env:
      NODE_ENV: production
    scale:
      replica: [2, 10]          # Auto-scaling range
      cpu: 256m
      memory: 512Mi
    ports:
      - 3000
      - "8080:3000"
    health:
      interval: 30
      command: "curl -f localhost:3000/health"
      retries: 3
      onfailure: restart        # or stop
    volumes:
      - path: /data
    provider:
      name: aws
      options:
        cluster: my-cluster
        namespace: production

Documentation

Full documentation available at https://arcaelas.github.io/pctl/

• Installation
• Getting Started
• Configuration
• AWS Provider
• GCP Provider
• Docker Provider
• API Reference

License

Custom License with Commercial Use Restrictions. See LICENSE for details.

Built by Arcaelas Insiders