Skip to main content
GitLab Auto DevOps is a built-in CI/CD feature that automatically detects your project’s language, framework, and configuration to generate a complete pipeline. It follows best practices across build, test, and deployment stages—integrating security scans, code quality checks, and container analysis—so you can focus on writing code instead of maintaining CI/CD scripts. Auto DevOps provides:
  • Automated build and containerization
  • Comprehensive testing (unit, integration, security)
  • Review Apps for testing merge request changes
  • Continuous delivery or deployment after merge
Auto DevOps offers sensible defaults, but you can customize the pipeline via CI/CD templates or project-specific .gitlab-ci.yml overrides.
The image is a flowchart illustrating the Auto DevOps process, detailing stages like continuous integration, packaging, and deployment with various automated testing and deployment steps.

​
How Auto DevOps Works at the Project Level

When you enable Auto DevOps in Settings > CI/CD, GitLab inspects your repository and applies predefined CI/CD templates. The default pipeline includes three core stages:

​
1. Build Stage

  • Detects a Dockerfile in your repo and builds a container image.
  • Falls back to Heroku buildpacks if no Dockerfile is found.
  • Outputs a ready-to-use Docker image for subsequent stages.

​
2. Test Stage

Runs your test suite and adds built-in checks: Supported languages include Ruby, Node.js, Java (Maven/Gradle), Python, Go, and more. All reports appear in the pipeline UI for immediate feedback.

​
3. Kubernetes Deployment

If you register a Kubernetes cluster in Operations > Kubernetes, Auto DevOps can deploy your app automatically. Supported cluster providers include:

​
Review Apps and Security Testing

When a merge request is opened, Auto DevOps spins up a Review App—a temporary, live environment to validate changes before merge. This deployment uses the Helm Auto Deploy chart, which you can customize. Once the Review App is live, Auto DevOps runs Dynamic Application Security Testing (DAST) using OWASP ZAP. ZAP crawls the application, identifies vulnerabilities, and produces a comprehensive report with severity levels and remediation advice.
The image shows an Auto DevOps pipeline with stages including Build, Test, Review, Dast, Production, and Performance, each containing various tasks.

​
Post-Merge Deployment and Performance Testing

After you merge to the default branch, Auto DevOps can deploy your application to staging or production based on your configuration. Post-deployment, it executes browser-based performance tests to benchmark page load times against previous releases—ensuring optimal user experience.

​
Deployment Strategies

Choose from three release workflows to match your team’s requirements:
Ensure your rollback procedures are tested and documented. Timed rollouts reduce risk but require proper monitoring and alerts.
The image illustrates an Auto DevOps Deployment Strategy, showing different deployment methods: continuous deployment to production, automatic deployment to staging with manual production deployment, and continuous deployment using timed incremental rollout.

Watch Video