Configuration: How do I use "defaults" to reduce the duplication of configuration?

In the following example the definition of the cluster and environment name variables are duplicated across tasks:

tasks:
    initialise:
        run:
            cluster: ${CLUSTER}
            environment:
                name:  '"sfdemo-" ~ code_reference.branch'
            # ...
   deployments:
        deploy:
            cluster: ${CLUSTER}
            environment:
                name:  '"sfdemo-" ~ code_reference.branch'
            # ...

You can avoid this duplication by defining them outside the tasks section in a separate defaults section:

defaults:
     cluster: ${CLUSTER}
     environment:
        name:  '"sfdemo-" ~ code_reference.branch'

tasks:
    initialise:
        run:
            # ...
    deployments:
        deploy:
            # ...

The cluster and environment name values will now be set for both the initialise and deployments tasks within the tasks section.