The Python Workers platform leverages FFI ↗ to access bindings to Cloudflare resources. Refer to the bindings documentation for more information.
From the configuration perspective, enabling Python Workflows requires adding the python_workflows compatibility flag to your Wrangler configuration file.
{
"$schema": "./node_modules/wrangler/config-schema.json",
"name": "workflows-starter",
"main": "src/index.py",
// Set this to today's date
"compatibility_date": "2026-07-20",
"compatibility_flags": ["python_workflows", "python_workers"],
"workflows": [
{
// name of your workflow
"name": "workflows-starter",
// binding name env.MY_WORKFLOW
"binding": "MY_WORKFLOW",
// this is class that extends the Workflow class in src/index.py
"class_name": "MyWorkflow",
}
]
}"$schema" = "./node_modules/wrangler/config-schema.json"
name = "workflows-starter"
main = "src/index.py"
# Set this to today's date
compatibility_date = "2026-07-20"
compatibility_flags = [ "python_workflows", "python_workers" ]
[[workflows]]
name = "workflows-starter"
binding = "MY_WORKFLOW"
class_name = "MyWorkflow"And this is how you use the payload in your workflow:
from workers import WorkflowEntrypoint
class DemoWorkflowClass(WorkflowEntrypoint):
async def run(self, event, step):
@step.do('step-name')
async def first_step():
payload = event["payload"]
return payloadThe Workflow binding gives you access to the Workflow class. All its methods are available
on the binding.
Create (trigger) a new instance of a given Workflow.
create(options=None)*options- an optional dictionary of options to pass to the workflow instance. Should contain the same keys as the WorkflowInstanceCreateOptions type.
from workers import WorkerEntrypoint, Response
class Default(WorkerEntrypoint):
async def fetch(self, request):
event = {"foo": "bar"}
await self.env.MY_WORKFLOW.create(params=event)
return Response.json({"status": "success"})The create method returns a WorkflowInstance object, which can be used to query the status of the workflow instance. Note that this is a Javascript object, and not a Python object.
Create (trigger) a batch of new workflow instances, up to 100 instances at a time. This is useful if you need to create multiple instances at once within the instance creation limit.
create_batch(batch)*batch- list ofWorkflowInstanceCreateOptionsto pass when creating an instance, including a user-provided ID and payload parameters.
Each element of the batch list is expected to include both id and params properties:
from workers import WorkerEntrypoint, Response
class Default(WorkerEntrypoint):
async def fetch(self, request):
# Create a new batch of 3 Workflow instances, each with its own ID and pass params to the Workflow instances
instances = [
{"id": "id-abc123", "params": {"hello": "world-0"}},
{"id": "id-def456", "params": {"hello": "world-1"}},
{"id": "id-ghi789", "params": {"hello": "world-2"}},
]
await self.env.MY_WORKFLOW.create_batch(instances)
return Response.json({"status": "success"})Get a workflow instance by ID.
get(id)*id- the ID of the workflow instance to get.
Returns a WorkflowInstance object, which can be used to query the status of the workflow instance.
from workers import WorkerEntrypoint, Response
class Default(WorkerEntrypoint):
async def fetch(self, request):
instance = await self.env.MY_WORKFLOW.get("abc-123")
# FFI methods available for WorkflowInstance
await instance.status()
await instance.pause()
await instance.resume()
await instance.restart()
await instance.terminate()
return Response.json({"status": "success"})Send an event to a workflow instance.
send_event(type, payload)*type- the type of event to send to the workflow instance. *payload- the payload to send to the workflow instance.
from workers import WorkerEntrypoint, Response
class Default(WorkerEntrypoint):
async def fetch(self, request):
await self.env.MY_WORKFLOW.send_event(type="my-event-type", payload={"foo": "bar"})
return Response.json({"status": "success"})Refer to the Workflows REST API documentation.
Refer to the CLI quick start to learn more about how to manage and trigger Workflows via the command-line.