Short answer: not directly. Assuming you’re on Prefect 3.x,

prefect.yaml

is a declarative project/deployment config and it does not execute arbitrary Python files. It’s used by

prefect deploy

to locate and import your flow entrypoint, set deployment options, and orchestrate build/push steps. It won’t create Blocks, Variables, or Automations by “running” Python for you. What you can do instead: - Blocks - Create/register block types with the CLI or Python once (e.g., in CI) and then reference them from your code/deployments. - CLI examples:

prefect block register -m your_package.blocks
    prefect block ls
    prefect block inspect <block_type>/<block_name>

- Docs: Blocks concept • prefect block CLI - Variables - Manage Prefect Variables via CLI or UI; reference them from your flow code. - CLI examples:

prefect variable set MY_VAR "some value"
    prefect variable get MY_VAR
    prefect variable ls

- Docs: prefect variable CLI - Automations - Define automations in the UI or via the API/SDK. The CLI can list/inspect/pause/resume/delete them, but does not create them. - CLI examples:

prefect automation ls
    prefect automation inspect <id>
    prefect automation pause <id>
    prefect automation resume <id>

- Docs: prefect automation CLI If you need to “run a Python script before/after deploy” (e.g., to create/update Blocks/Variables), do that as a separate step in your CI pipeline or local script before calling

prefect deploy

.

prefect deploy

will import your flow entrypoint (e.g.,

./path/to/flow.py:flow_func

), but relying on import-time side effects to create resources is not recommended. Reference for

prefect.yaml

structure and capabilities: prefect.yaml guide If you can share what you want to automate (e.g., “create an S3 block, set two variables, then deploy this flow”), I can suggest a concrete, repeatable workflow for it.