Skip to main content

Publish anything to the Artifacts tab

You can use the Artifact Metadata Publisher plugin to publish any URL on the Artifacts tab of the Build details page.

tip

If you use AWS S3, you can use either the Artifact Metadata Publisher plugin or the S3 Upload and Publish plugin, which combines the upload artifact and publish URL steps in one plugin.

Configure the Artifact Metadata Publisher plugin​

To use the Artifact Metadata Publisher plugin, add a Plugin step to your CI pipeline.

For artifacts generated in the same pipeline, the Plugin step is usually placed after the step that uploads the artifact to cloud storage.

- step:
    type: Plugin
    name: publish artifact metadata
    identifier: publish_artifact_metadata
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: plugins/artifact-metadata-publisher
      settings:
        file_urls: https://domain.com/path/to/artifact

Plugin step specifications​

Use these specifications to configure the Plugin step to use the Artifact Metadata Publisher plugin.

connectorRef​

Use the built-in Docker connector (account.harness.Image) or specify your own Docker connector. Harness uses this to pull the plugin Docker image.

This setting is labeled Container Registry in the Visual editor.

image​

Set to plugins/artifact-metadata-publisher.

file_urls​

Provide the URL for the artifact you want to link on the Artifacts tab. In the Visual editor, set this as a key-value pair under Settings.

For artifacts in cloud storage, use the appropriate URL format for your cloud storage provider, for example:

  • GCS: https://storage.googleapis.com/GCS_BUCKET_NAME/TARGET_PATH/ARTIFACT_NAME_WITH_EXTENSION
  • S3: https://BUCKET.s3.REGION.amazonaws.com/TARGET/ARTIFACT_NAME_WITH_EXTENSION

For artifacts uploaded through Upload Artifacts steps, use the Bucket, Target, and artifact name specified in the Upload Artifacts step.

For private S3 buckets, use the console view URL, such as https://s3.console.aws.amazon.com/s3/object/BUCKET?region=REGION&prefix=TARGET/ARTIFACT_NAME_WITH_EXTENSION.

If you uploaded multiple artifacts, you can provide a list of URLs, such as:

file_urls:
  - https://BUCKET.s3.REGION.amazonaws.com/TARGET/artifact1.html
  - https://BUCKET.s3.REGION.amazonaws.com/TARGET/artifact2.pdf

You can also set display name for each of the URLs set in the plugin, by using the format name:::file_url, where ::: is the delimiter that separates the name and url. For example:

file_urls:
  - artifact1:::https://BUCKET.s3.REGION.amazonaws.com/TARGET/artifact1.html
  - artifact2:::https://BUCKET.s3.REGION.amazonaws.com/TARGET/artifact2.pdf

For information about using Harness CI to upload artifacts, go to Build and push artifacts and images.

Artifact metadata publisher environment variables​

The artifact-metadata-publisher plugin looks for two environment variables:

  • PLUGIN_ARTIFACT_FILE: The file path where the plugin stores artifact metadata
  • PLUGIN_FILE_URLS: The URLs to be published to the Artifacts tab

The URLs passed via PLUGIN_FILE_URLS are stored in PLUGIN_ARTIFACT_FILE in the following JSON format, which the CI manager expects to publish under the Artifacts tab:

{
  "kind": "fileUpload/v1",
  "data": {
    "fileArtifacts": [
      {
        "name": "file-0",
        "url": "https://example.com/artifact"
      }
    ]
  }
}

Kubernetes flow differences​

In Kubernetes flow, the CI manager doesn't automatically create the temporary file and pass it via PLUGIN_ARTIFACT_FILE (unlike cloud flow). You need to pass artifact_file: somefile.txt explicitly as input. The plugin then creates this file (somefile.txt) in the JSON format above, containing the URLs passed via PLUGIN_FILE_URLS input, which is then processed and displayed under the Artifacts tab in the pipeline.

- step:
    type: Plugin
    name: publish artifact metadata
    identifier: publish_artifact_metadata
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: plugins/artifact-metadata-publisher
      settings:
        file_urls: https://domain.com/path/to/artifact
        artifact_file: temp.txt

Using the plugin binary on hosted macOS runners​

Hosted macOS runners cannot run Docker containers, which means you cannot use the containerized Plugin step with the artifact-metadata-publisher. Instead, you can download and run the plugin binary directly in a Run step.

Download and use the plugin binary​

Use the following commands in a Run step to download the appropriate binary for macOS and publish artifacts:

- step:
    type: Run
    name: Publish Artifact Metadata
    identifier: publish_artifact_metadata
    spec:
      shell: Sh
      command: |-
        # Download the artifact-metadata-publisher binary for macOS ARM64
        curl -L https://github.com/drone-plugins/artifact-metadata-publisher/releases/download/v2.2.0/artifact-metadata-publisher-darwin-arm64.zst -o artifact-metadata-publisher-darwin-arm64.zst
        
        # Decompress the binary
        zstd -d artifact-metadata-publisher-darwin-arm64.zst -o ./artifact-metadata-publisher
        
        # Make the binary executable
        chmod 700 ./artifact-metadata-publisher
        
        # Run the plugin with your artifact URLs
        PLUGIN_FILE_URLS="https://domain.com/path/to/artifact" ./artifact-metadata-publisher

Note​

  • Replace v2.2.0 with the latest version from the artifact-metadata-publisher releases.
  • For macOS Intel (x86_64), use artifact-metadata-publisher-darwin-amd64.zst instead.
  • The CI manager automatically creates a temporary file and injects it via the PLUGIN_ARTIFACT_FILE environment variable, so you don't need to create or manage this file manually.

Publish multiple artifacts with custom names​

You can publish multiple artifacts and set custom display names using the format name:::url, where ::: is the delimiter between the name and URL:

- step:
    type: Run
    name: Publish Multiple Artifacts
    identifier: publish_multiple_artifacts
    spec:
      shell: Sh
      command: |-
        curl -L https://github.com/drone-plugins/artifact-metadata-publisher/releases/download/v2.2.0/artifact-metadata-publisher-darwin-arm64.zst -o artifact-metadata-publisher-darwin-arm64.zst
        zstd -d artifact-metadata-publisher-darwin-arm64.zst -o ./artifact-metadata-publisher
        chmod 700 ./artifact-metadata-publisher
        
        # Publish multiple artifacts with custom names
        PLUGIN_FILE_URLS="Config File:::https://s3.amazonaws.com/bucket/config.json,Build Report:::https://s3.amazonaws.com/bucket/report.html" ./artifact-metadata-publisher

Example: Publishing S3 presigned URLs​

Here's a complete example that generates a presigned URL for an S3 artifact and publishes it to the Artifacts tab:

- step:
    type: Run
    name: Generate Presigned URL
    identifier: generate_presigned_url
    spec:
      shell: Sh
      command: |-
        FILE_NAME="my-artifact.zip"
        
        # Generate presigned URL (assumes AWS credentials are configured)
        PRESIGNED_URL=$(aws s3 presign s3://my-bucket/$FILE_NAME --expires-in 3600)
        
        echo "Generated presigned URL: $PRESIGNED_URL"
      envVariables:
        AWS_ACCESS_KEY_ID: <+secrets.getValue("aws_access_key")>
        AWS_SECRET_ACCESS_KEY: <+secrets.getValue("aws_secret_key")>
      outputVariables:
        - name: PRESIGNED_URL

- step:
    type: Run
    name: Publish Artifact Metadata
    identifier: publish_artifact_metadata
    spec:
      shell: Sh
      command: |-
        curl -L https://github.com/drone-plugins/artifact-metadata-publisher/releases/download/v2.2.0/artifact-metadata-publisher-darwin-arm64.zst -o artifact-metadata-publisher-darwin-arm64.zst
        zstd -d artifact-metadata-publisher-darwin-arm64.zst -o ./artifact-metadata-publisher
        chmod 700 ./artifact-metadata-publisher
        
        # Use the presigned URL from the previous step
        PLUGIN_FILE_URLS="my-artifact.zip:::<+execution.steps.generate_presigned_url.output.outputVariables.PRESIGNED_URL>" ./artifact-metadata-publisher

Build logs and artifact files​

When you run the pipeline, you can observe the step logs on the Build details page, and you can find the artifact URL on the Artifacts tab.

On the Artifacts tab, select the step name to expand the list of artifact links associated with that step.

If your pipeline has multiple steps that upload artifacts, use the dropdown menu on the Artifacts tab to switch between lists of artifacts uploaded by different steps.

Tutorial: Upload an Allure report to the Artifacts tab​

This tutorial demonstrates how to use the Artifacts Metadata Publisher plugin by generating and uploading an Allure report.

Part 1: Prepare cloud storage

  1. You need access to a cloud storage provider. This tutorial uses GCS. If you use another option, you'll need to modify some of the steps according to your chosen cloud storage provider.

  2. If you use S3, GCS, or JFrog, you need a Harness connector to use with the Upload Artifact step:

    tip

    Store authentication and access keys for connectors as Harness secrets.

  3. In your cloud storage, create a bucket or repo where you can upload your artifact.

    To access the artifact directly from the Artifacts tab, the upload location must be publicly available. If the location is not publicly available, you might need to log in to view the artifact or use a different artifact URL (such as a console view URL). This tutorial uses a publicly available GCS bucket to store the report.

Part 2: Prepare artifacts to upload

Add steps to your pipeline that generate and prepare artifacts to upload, such as Run steps. The steps you use depend on what artifacts you ultimately want to upload.

For example, this tutorial uses three Run steps to generate and prepare an artifact:

  • The first step runs tests with Maven.
  • The second step generates an Allure report. To ensure the build environment has the Allure tool, the step uses a Docker image that has this tool: solutis/allure:2.9.0.
  • The third step combines the Allure report into a single HTML file.
    • To view an Allure report in a browser, you must run a web server with the allure open command; however, this command won't persist after the CI pipeline ends. Instead, use the allure-combine tool to convert the Allure report into a single HTML file.
    • Running the allure-combine . command inside allure-report generates the complete.html file.
    • To ensure the build environment as access to the allure-combine tool, you can include steps to install it or use a Docker image that has the tool, such as shubham149/allure-combine:latest.
- step:
    type: Run
    name: run maven tests
    identifier: run_maven_tests
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: openjdk:11
      shell: Sh
      command: ./mvnw clean test site
- step:
    type: Run
    name: generate allure report
    identifier: generate_allure_report
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: solutis/allure:2.9.0
      command: |
        cd target
        allure generate allure-results --clean -o allure-report
- step:
    type: Run
    name: combine report
    identifier: combine_report
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: shubham149/allure-combine:latest
      command: |
        cd target/allure-report
        allure-combine .
        cd ../..
        cp target/allure-report/complete.html .
Part 3: Upload to cloud storage

Add a step to upload your artifact to cloud storage:

For example, this tutorial uploads the combined Allure report to GCS:

- step:
    type: GCSUpload
    name: upload report
    identifier: upload_report
    spec:
      connectorRef: YOUR_GCP_CONNECTOR_ID
      bucket: YOUR_GCS_BUCKET
      sourcePath: target/allure-report/complete.html
      target: <+pipeline.sequenceId>
tip

The target value uses a Harness expression, <+pipeline.sequenceId>, to ensure that artifacts uploaded by this pipeline are stored in unique directories and don't overwrite one another.

Part 4: Add URLs to the Artifacts tab

At this point, you can run the pipeline and then manually find the uploaded artifact in your cloud storage bucket or repo. Alternately, you can use the Artifact Metadata Publisher plugin, which makes it easier to find the artifact directly associated with a particular build.

For example, this step publishes the URL for the combined Allure report on the Artifacts tab:

- step:
    type: Plugin
    name: publish artifact metadata
    identifier: publish_artifact_metadata
    spec:
      connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
      image: plugins/artifact-metadata-publisher
      settings:
        file_urls: https://storage.googleapis.com/YOUR_GCS_BUCKET/<+pipeline.sequenceId>/complete.html

For details about these settings, go to Configure the Artifact Metadata Publisher plugin.

Tutorial YAML examples

These YAML examples demonstrate the pipeline created in the preceding tutorial. This pipeline:

  1. Builds a Java Maven application.
  2. Generates and compiles an Allure report.
  3. Uploads the report to cloud storage.
  4. Provides a URL to access the report from the Artifacts tab in Harness.

This example uses Harness Cloud build infrastructure and uploads the Allure report artifact to GCS.

pipeline:
  name: allure-report-upload
  identifier: allurereportupload
  projectIdentifier: YOUR_HARNESS_PROJECT_ID
  orgIdentifier: default
  tags: {}
  properties:
    ci:
      codebase:
        connectorRef: YOUR_CODEBASE_CONNECTOR_ID
        repoName: YOUR_CODE_REPO_NAME
        build: <+input>
  stages:
    - stage:
        name: test and upload artifact
        identifier: test_and_upload_artifact
        description: ""
        type: CI
        spec:
          cloneCodebase: true
          platform:
            os: Linux
            arch: Amd64
          runtime:
            type: Cloud
            spec: {}
          execution:
            steps:
              - step:
                  type: Run
                  name: run maven tests
                  identifier: run_maven_tests
                  spec:
                    connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
                    image: openjdk:11
                    shell: Sh
                    command: ./mvnw clean test site
              - step:
                  type: Run
                  name: generate allure report
                  identifier: generate_allure_report
                  spec:
                    connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
                    image: solutis/allure:2.9.0
                    command: |
                      cd target
                      allure generate allure-results --clean -o allure-report
              - step:
                  type: Run
                  name: combine report
                  identifier: combine_report
                  spec:
                    connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
                    image: shubham149/allure-combine:latest
                    command: |
                      cd target/allure-report
                      allure-combine .
                      cd ../..
                      cp target/allure-report/complete.html .
              - step:
                  type: GCSUpload
                  name: upload report
                  identifier: upload_report
                  spec:
                    connectorRef: YOUR_GCP_CONNECTOR_ID
                    bucket: YOUR_GCS_BUCKET
                    sourcePath: target/allure-report/complete.html
                    target: <+pipeline.sequenceId>
              - step:
                  type: Plugin
                  name: publish artifact metadata
                  identifier: publish_artifact_metadata
                  spec:
                    connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
                    image: plugins/artifact-metadata-publisher
                    settings:
                      file_urls: https://storage.googleapis.com/YOUR_GCS_BUCKET/<+pipeline.sequenceId>/complete.html

See also​

Troubleshoot the Artifacts tab​

Go to the CI Knowledge Base for questions and issues related to the Artifacts tab, such as: