Based on the research and signature verification, here's the comprehensive answer: Prefect variables cannot be directly referenced within flow decorators themselves. The flow decorator accepts specific parameters (as shown in the signature above) that are set at definition time, not runtime. However, there are two important related capabilities: 1. You can use Prefect variables within the flow function itself:

python
from prefect import flow, Variable

@flow
def my_flow():
    my_var = Variable.get("my-variable")
    print(f"The variable value is: {my_var}")

2. You can reference variables in your deployment configuration (prefect.yaml) for deployment-time customization:

yaml
deployments:
  - name: my-deployment
    flow_name: my-flow
    parameters:
      param1: "{{ prefect.variables.MY_VARIABLE }}"

If you need dynamic behavior that would typically be set in a decorator, the recommended approach is to: 1. Use deployment configurations to set variables 2. Access those variables within your flow using

Variable.get()

3. Use control flow within your function based on those variables Is there a specific use case you're trying to achieve with variables in decorators? That might help me provide more targeted guidance.