@Marvin - how can I define environment variables which should get set in the flow run environment from the deployment yaml file

To set environment variables for a flow run in a deployment YAML file, you can use the

job_variables

section to provide these variables explicitly. Here's a brief overview of how you might define these in your

prefect.yaml

file for both Prefect 2.x and 3.x. For Prefect 2.x, you can define environment variables under

job_variables

. Here's an example snippet:

yaml
deployments:
- name: demo-deployment
  entrypoint: path/to/your_flow.py:flow_name
  work_pool:
    name: local
  job_variables:
    env:
      EXECUTION_ENVIRONMENT: staging
      MY_NOT_SO_SECRET_CONFIG: plumbus
  schedule: null

Source for Prefect 2.x For Prefect 3.x, the configuration in

prefect.yaml

allows referencing environment variables dynamically, which can be useful for runtime settings:

yaml
{{ $MY_ENV_VAR }}

This structure enables setting environment variables that can be utilized during the execution. This approach leverages Prefect's infrastructure features to dynamically assign values to variables, such as an image name and tag for a Docker deployment. Source for Prefect 3.x If only life could be as consistently version-controlled as YAML files, alas, it deals in much more mysterious variables...