This documentation is for the Technical Preview of JFrog Pipelines, available exclusively to JFrog Enterprise+ customers

Skip to end of metadata
Go to start of metadata


Once JFrog Pipelines is installed, you can get up and running easily through its interactive UI. After going through this page, you should have an understanding of the fundamental concepts and tasks of creating and running a pipeline workflow in JFrog Pipelines.


Before learning how to use Pipelines, here are some fundamental concepts you will need to be familiar with.


An Integration connects JFrog Pipelines to an external service/tool. Each integration type defines the endpoint, credentials and any other configuration detail required for Steps in a Pipeline to interact with the source. All credential information is encrypted and maintained separately from the pipeline definition, and held in a secure storage in conformance with best practices.

For more information, and a list of all available integration types, see the Integrations reference.


Resources typically contain information needed for a step in a pipeline to execute and can also be used to store information produced by a step. Resources defined should be unique across all the pipelines. Because of that they also provide the way to link pipelines. 

For more information, and a list of all available resource types, see the Resources reference.


step is an unit of execution in a pipeline. It is triggered by some event and uses resources to performs a build action as part of the pipeline workflow.

For more information, and a list of all available step types, see the Steps reference.


pipeline is a workflow composed of a series of event-driven, interconnected steps to execute serially or in parallel to achieve an outcome.


run is an instance of execution of a pipeline.

Accessing Pipelines Using Single Sign-On (SSO)

Use your Authentication Provider Artifactory login credentials to log in to JFrog Pipelines.

Contact your administrator if you do not know your credentials.

Configuring Pipelines

When JFrog Pipelines is first installed, you'll need to configure it so that your workflows have places to execute and can connect to the facilities they will use. All of these activities are started through the Configuration button.

  • Create at least one node pool of an architecture/OS and populate it with nodes (virtual machines) where the steps of your pipeline can execute.
  • Add the integrations you'll be needing to access other facilities from Pipelines, such as for your GitHub account, and Artifactory.
  • Add at least one pipeline source (such as your GitHub repo) where you will be keeping your YAML-based pipeline config files. You can add as many repositories as you will use, and each user can add their own as they need them at any time in future.

Define a Pipeline

To create a pipeline workflow, you declare the resources you will use and specify the steps to execute in one or more YAML files. When those YAML files are stored in one of the pipeline sources you configured, Pipelines automatically loads them and waits for a triggering event to start dispatching the pipeline's steps to nodes for execution.

For high-level understanding of the structure of these YAML files, review Defining a Pipeline.

Just to get started, take a look at these example YAML files for pipelines:

View and Run Pipelines

Once you've created your pipeline YAML file, you must commit it to the repository you have set up as a pipeline source. Pipelines will then automatically load it, and sync when any changes to it are made. 

Then you can see all of the pipelines that were loaded from all pipeline sources. 

Clicking on your pipeline's name in the list will display it in an interactive visualization, along with the run history of the pipeline.

Read Viewing Pipelines to learn how to navigate these views.

To test your pipeline, you can trigger manually through the interactive visualization, starting from any step in the pipeline sequence.

Read Running a Pipeline to learn how to trigger your pipeline and view what it executes.

  • No labels