workflows-py
workflows-py copied to clipboard
Workflows are an event-driven, async-first, step-based way to control the execution flow of AI applications like agents.
Initial stab at attaching arbitrary metadata to handlers - metadata is assigned when calling run() - if continuing a run/session, it will re-use existing metadata (or overwrite) - the list...
### Summary This PR integrates [DBOS](https://github.com/dbos-inc/dbos-transact-py) with workflow and context to provide out-of-the-box durable execution and checkpointing. ### Changes * **DBOSWorkflow:** * It subclasses `Workflow` to start a DBOS durable...
Currently, to do a map reduce pattern, you need a decent amount of boilerplate ```py import random from workflows import Context, Workflow, step from workflows.events import Event, StartEvent, StopEvent class...
### Feature Description - Currently as the doc says draw_most_recent_execution, draw_most_recent_execution do not draw nested workflows steps but this should be super awesome! - I think we can save the...
### Bug Description I'm encountering a persistent *WorkflowValidationError* when defining a custom Workflow class in LlamaIndex. The error occurs during class definition when using the *@step* decorator on methods that...
### Feature Description Implement an exception handler that includes all step context, allowing for user customization. ### Reason Using workflows we can use retry_policy, but when max retries on a...
Today, each workflow is completely decoupled from another. And furthermore, the `Context` ensures each `.run()` is also isolated. If you run a workflow inside another workflow, you end up with...
Are there any plans to support persistent state stores? Currently it looks like `InMemoryStateStore` is hardcoded in the `Context`.