GitHub Actions allow you to customize your workflows to meet the unique needs of your application and team. In this guide, we'll discuss some of the essential customization techniques such as using variables, running scripts, and sharing data and artifacts between jobs.
Using variables in your workflows
GitHub Actions include default environment variables for each workflow run. If you need to use custom environment variables, you can set these in your YAML workflow file. This example demonstrates how to create custom variables named
POSTGRES_PORT. These variables are then available to the
node client.js script.
jobs: example-job: steps: - name: Connect to PostgreSQL run: node client.js env: POSTGRES_HOST: postgres POSTGRES_PORT: 5432
For more information, see "Variables."
Adding scripts to your workflow
You can use actions to run scripts and shell commands, which are then executed on the assigned runner. This example demonstrates how an action can use the
run keyword to execute
npm install -g bats on the runner.
jobs: example-job: steps: - run: npm install -g bats
For example, to run a script as an action, you can store the script in your repository and supply the path and shell type.
jobs: example-job: steps: - name: Run build script run: ./.github/scripts/build.sh shell: bash
For more information, see "Workflow syntax for GitHub Actions."
Sharing data between jobs
If your job generates files that you want to share with another job in the same workflow, or if you want to save the files for later reference, you can store them in GitHub as artifacts. Artifacts are the files created when you build and test your code. For example, artifacts might include binary or package files, test results, screenshots, or log files. Artifacts are associated with the workflow run where they were created and can be used by another job. All actions and workflows called within a run have write access to that run's artifacts.
For example, you can create a file and then upload it as an artifact.
jobs: example-job: name: Save output steps: - shell: bash run: | expr 1 + 1 > output.log - name: Upload output file uses: actions/upload-artifact@v3 with: name: output-log-file path: output.log
To download an artifact from a separate workflow run, you can use the
actions/download-artifact action. For example, you can download the artifact named
jobs: example-job: steps: - name: Download a single artifact uses: actions/download-artifact@v3 with: name: output-log-file
To download an artifact from the same workflow run, your download job should specify
needs: upload-job-name so it doesn't start until the upload job finishes.
For more information about artifacts, see "Storing workflow data as artifacts."
To continue learning about GitHub Actions, see "About workflows."