In a task (Prefect 3.x), the simplest way is to use the runtime accessors:

from prefect import flow, task
from prefect.runtime import flow_run

@task
def show_flow_run_name():
    print(f"Flow run name: {flow_run.name}")

@flow
def my_flow():
    show_flow_run_name()

if __name__ == "__main__":
    my_flow()

Alternative: use the run context + client (useful if you need more than just the name). In a task,

get_run_context()

returns a

TaskRunContext

, so you’ll fetch the flow run by ID: - Async task

from prefect import flow, task
from prefect.context import get_run_context
from prefect.client.orchestration import get_client

@task
async def show_flow_run_name():
    ctx = get_run_context()  # TaskRunContext
    async with get_client() as client:
        fr = await client.read_flow_run(ctx.task_run.flow_run_id)
        print(f"Flow run name: {fr.name}")

@flow
def my_flow():
    show_flow_run_name()

if __name__ == "__main__":
    my_flow()

- Sync task

from prefect import flow, task
from prefect.context import get_run_context
from prefect.client.orchestration import get_client

@task
def show_flow_run_name():
    ctx = get_run_context()  # TaskRunContext
    with get_client(sync_client=True) as client:
        fr = client.read_flow_run(ctx.task_run.flow_run_id)
        print(f"Flow run name: {fr.name}")

@flow
def my_flow():
    show_flow_run_name()

if __name__ == "__main__":
    my_flow()

Notes and caveats: -

prefect.runtime.flow_run.name

is the easiest and works inside tasks; if there’s no active run it may be

None

, so guard if needed. - In a task,

get_run_context()

returns a

TaskRunContext

which does not include the flow run object; use

ctx.task_run.flow_run_id

+ client to fetch details. - Prefer

prefect.runtime.flow_run.name

if you only need the name. Docs: - Access runtime info (runtime.*) - Runtime and run context concepts - get_run_context API