Skip to main content
Skyhook provides automated CI/CD pipelines for your services using GitHub Actions. When you create or import a service, Skyhook generates ready-to-use workflows that handle building Docker images, deploying to Kubernetes, and managing your entire release process.

What You Get

Skyhook’s CI/CD system is built on composable GitHub Actions workflows that handle:
  • Automated Builds - Docker images built and pushed to your container registry
  • Multi-Environment Deployments - Deploy to dev, staging, production, and custom environments
  • GitOps Integration - Optional ArgoCD support for declarative deployments
  • Progressive Delivery - Argo Rollouts for canary and blue-green deployments
  • Multi-Cloud Support - Works with AWS (EKS + ECR), GCP (GKE + GCR), and Azure (AKS + ACR)
  • Monorepo Support - Build and deploy multiple services from a single repository
All workflows are customizable and can be modified to fit your specific needs.

How It Works

When you create a service with Skyhook, we automatically generate GitHub Actions workflows in your repository:
.github/
└── workflows/
    ├── release.yml          # Automated CI/CD on push to main
    ├── build_and_deploy.yml # Manual build + deploy
    ├── deploy.yml           # Manual deploy only
    └── build_image.yml      # Manual build only
These workflows integrate with:
  • Your container registry (ECR, GCR, ACR, Docker Hub, GHCR)
  • Your Kubernetes clusters
  • Your Git repository (for GitOps deployments)

Deployment Methods

Skyhook supports two deployment approaches that you can choose based on your needs:

GitOps with ArgoCD

How it works:
  1. Workflow updates Kubernetes manifests in your Git repository
  2. Commits the changes automatically
  3. ArgoCD detects the change and syncs to your cluster
  4. Provides automatic drift detection and self-healing
Best for:
  • Production environments
  • Teams requiring audit trails
  • Organizations with strict compliance requirements
Benefits:
  • Git as single source of truth
  • Complete audit trail of all changes
  • Easy rollbacks via Git history
  • Declarative infrastructure
  • Automatic drift detection
Learn more about GitOps →

Direct Deployment (kubectl)

How it works:
  1. Workflow authenticates directly to your Kubernetes cluster
  2. Applies manifests using kubectl
  3. Optionally waits for Argo Rollouts progressive delivery to complete
Best for:
  • Development and staging environments
  • Smaller teams
  • Rapid iteration
Benefits:
  • Faster deployment (no Git commit step)
  • Simpler setup
  • Direct feedback in CI/CD logs
  • No additional infrastructure needed

Workflow Types

Automated Workflows

Release Workflow - Triggers automatically on every push to your main branch:
  • Builds Docker images with auto-generated tags
  • Pushes to your container registry
  • Optionally deploys to development environment
  • Supports monorepo detection for building only changed services

Manual Workflows

Build and Deploy - Build a new image and deploy it in one operation:
  • Deploy from any branch, tag, or commit
  • Test feature branches before merging
  • Create releases from specific versions
Deploy Existing Image - Deploy an already-built image:
  • Promote images from staging to production
  • Rollback to previous versions
  • Deploy the same tested image to multiple environments
Build Image Only - Build and push without deploying:
  • Pre-build images for later deployment
  • Create release candidates
  • Test Docker builds
Learn how to use workflows →

Multi-Cloud Support

Skyhook workflows work seamlessly across cloud providers with unified configuration:
  • AWS: EKS clusters + ECR registries
  • GCP: GKE clusters + Artifact Registry/GCR
  • Azure: AKS clusters + Azure Container Registry
Each cloud provider uses modern authentication (OIDC/Workload Identity) to eliminate long-lived credentials. Configure multi-cloud setup →

Getting Started

  1. Create or import a service - Skyhook generates workflows automatically
  2. Configure secrets - Set up cloud provider credentials in GitHub
  3. Push to main - The release workflow builds and deploys automatically
  4. Deploy manually - Use the Skyhook UI to deploy to other environments
Ready to deploy? Continue to: