Python: [Issue]: Clarification on ExperimentalWarning when using WorkflowFactory from agent-framework-declarative #10092

Category

Documentation

Problem

Description

We are using the agent-framework-declarative package to load and run declarative workflows using WorkflowFactory.in python

While running that, we are encountering the following warning:

ExperimentalWarning: [DECLARATIVE_AGENTS] AgentFactory is experimental and may change or be removed in future versions without notice.

Based on our understanding, the package provides two different stability levels:

Declarative workflows (WorkflowFactory, executors, handlers, and the _workflows surface) are stable.
Declarative agents (AgentFactory and the YAML agent loading/parsing path: DeclarativeLoaderError, ProviderLookupError, ProviderTypeMapping) are experimental and may change or be removed in future versions without notice. Using any of these symbols emits an ExperimentalWarning on first use.

Our implementation uses WorkflowFactory to load and run declarative workflows. However, internally, WorkflowFactory appears to utilize AgentFactory, which results in the experimental warning even though we are primarily using the stable workflow APIs.

Questions

1. Is there a plan or expected timeline for making AgentFactory stable?
2. Is the experimental warning expected when using WorkflowFactory, even though declarative workflows are documented as stable?
3. Is there any recommended alternative or workaround to avoid relying on the experimental AgentFactory functionality when running declarative workflows?
4. Will the internal dependency between WorkflowFactory and AgentFactory be removed or redesigned in a future release?

Evidence

After running below code
from agent_framework_declarative import WorkflowFactory

factory = WorkflowFactory()

workflow = factory.create_workflow_from_yaml("""
name: hello-workflow
actions:

• kind: SendActivity
  activity:
  text: ="Hello, " & inputs.name
  """)

Getting below warning
ExperimentalWarning: [DECLARATIVE_AGENTS] AgentFactory is experimental and may change or be removed in future versions without notice.

Desired Outcome

We would appreciate clarification on the stability of this internal dependency, the expected timeline for a stable release, and the recommended approach for applications using WorkflowFactory in production.

Thanks for your support!

Language

Python

Acknowledgements

• I searched existing issues and did not find a duplicate.
• I will wait for explicit maintainer agreement before starting implementation of a non-trivial change.