Pipelines
Overview of Harness pipelines, how to navigate to them, the pipelines and executions listing views, and two ways to create a pipeline.
A pipeline is the top-level execution unit in Harness. It contains one or more stages, and each stage groups steps that perform work: building an artifact, deploying a service, running tests, or waiting for approval. Pipelines can be triggered manually, by a webhook, on a schedule, or on a new artifact.
Navigate to Pipelines
Pipelines are accessible from anywhere in your project:
In the left navigation, select more to open the navigation panel.
Under Resources, select Pipelines. It is the first item in the resources list.
You can also search for "Pipelines" in the search field at the top of the panel.
Pipelines listing page
The Pipelines listing page shows all pipelines and their execution history. It has two tabs: Pipelines and Executions.
Pipelines tab
The Pipelines tab lists every pipeline in the project. The default view shows these columns:
Name
The pipeline name. Select it to open the pipeline in Pipeline Studio.
Store
Where the pipeline YAML is stored. Inline means stored within Harness; otherwise shows the Git repository.
Execution status
The status of the most recent execution (Success, Failed, Running, etc.).
Recent executions
A sparkline showing the last several execution results at a glance.
Created
When the pipeline was created.
Updated
When the pipeline was last modified.
All available columns (11 total): Name, Store, ID, Description, Tags, Version, Last executed, Execution status, Recent executions, Created, Updated. Select Columns to show or hide columns.
Filter and sort pipelines
Use these controls to narrow or reorder the list:
Search: Filter by pipeline name.
Add filter: Filter by Repository, Deployment type, Tags, Name, or Description.
Sort: Last executed (default), Last updated, Name (A to Z), Name (Z to A).
Items per page: Choose 10, 25, or 50 pipelines per page.
Each pipeline row has a ⋮ menu with actions such as run, edit, and delete.
Executions tab
The Executions tab lists all pipeline runs across the project, regardless of which pipeline triggered them. The default view shows 8 of 15 available columns. Select Columns to customize what is displayed.
ID
Unique identifier for the execution.
Pipeline name
The name of the pipeline that ran.
Tags
Tags applied to the execution.
Version
The pipeline version.
Repository
Source code repository associated with the run.
Commit
The commit SHA that triggered the run.
Committer
Who made the triggering commit.
Build artifacts
Artifacts produced by the run.
Deployment
Services and environments targeted by the run.
Input sets
Input sets used at runtime.
Notes
Custom notes added to the execution.
Trigger
What triggered the run: manual, webhook, schedule, or artifact.
Start time
When the execution started.
Duration
How long the execution took.
Pipeline timeout
The timeout configured for the pipeline.
Filter executions
Use these controls to narrow down the list:
Status: Select Status to filter by execution state. Available states: Success, Failed, Aborted, Running, Queued, Paused, Expired, Waiting, Approval Waiting, Intervention Waiting.
My executions: Select My executions to show only runs triggered by your user account.
Add filter: Select + Add filter to filter by Generic fields (Timeframe, Pipeline, Pipeline Tags, Execution Notes, Execution ID, Trigger Type, Trigger Identifier, Input Sets, Repository), Deployment fields (Deployment Type, Services, Environments, Artifacts, Build Repository, Branch), or Build fields (Build Repository, Build Type, Build Tag, Source Branch, Target Branch, PR Number, Commit Branch).
Sort executions
Select the sort dropdown to order by: Name (A to Z), Name (Z to A), Status (A to Z), Status (Z to A), Start time (newest first), Start time (oldest first), Pipeline timeout (newest first), Pipeline timeout (oldest first).
Create a pipeline
There are two ways to create a pipeline.
Visual editor
Build in Pipeline Studio
Harness opens Pipeline Studio. Build your pipeline in the Visual editor by adding stages and steps, or switch to the YAML editor to write the pipeline definition directly.
Go to Pipeline Studio for a full walkthrough.
Harness AI
You can describe a pipeline in natural language and let Harness AI create it for you. Select the Harness AI chat panel from the left side of any page in Harness to get started.
Open in Harness, review and run pipeline
After accepting, Harness AI confirms the pipeline is created and shows an Open in Harness button. Select it to open the pipeline in Pipeline Studio, where you can review, adjust, and run it.
AI-created pipelines appear immediately in the Pipelines listing page.
Stages
Harness does not lock a stage to a type — a stage is a container for steps, and you can combine any mix of steps in a single stage: build, deploy, test, or approve. When you add a stage, types such as Build or Deploy appear as starting points that pre-populate relevant configuration, but they do not restrict what you configure afterward.
Example
Select the Build template to configure CI steps and infrastructure. Then, in the same stage, add a service, an environment, and deployment steps right after your build steps. Build and deploy run together in a single stage.
Next steps
Go to Pipeline Studio to build and edit pipelines using the Visual and YAML editors.
Go to Pipeline settings to configure delegates, inputs, notifications, clone behavior, timeout, and execution controls.
Go to Stages to understand the stage model and available stage types.
Go to Steps to review the step types available inside a stage.
Go to Execution controls to run stages multiple times using looping strategies.
Go to Input sets to save and reuse runtime input values across runs.
Go to Execution logs to inspect logs and artifacts from a pipeline run.
Go to YAML reference to find every supported pipeline YAML key.
Last updated
Was this helpful?