@Scout I started to get this warning from a linter:

Metaclass conflict: the metaclass of a derived class must be a (non-strict) subclass of the metaclasses of all its bases  [misc]

for example for this code: class MyConfig(dagster.Config): some_field: str What's wrong?

It appears that you are encountering a metaclass conflict when using the

dagster.Config

class as a base for your

MyConfig

class. This issue has been reported by another user in a Github issue, where they experienced a similar problem with mypy type checking and the

dagster.Config

class after updating to Dagster version 1.5.5. The issue stems from a metaclass conflict between the

dagster.Config

class and the base classes it inherits from. Mypy, which is a static type checker for Python, is raising an error because the metaclass of

MyConfig

is not compatible with the metaclasses of its base classes. The user who reported the issue mentioned that while Pyright, another static type checker, did not raise any errors, mypy did. This suggests that the problem may be specific to how mypy interprets the metaclass hierarchy. Unfortunately, the documentation does not provide a direct solution to this problem. However, the user who reported the issue was looking for a workaround or fix that could be applied within the Dagster codebase to resolve the issue for mypy users. As a developer relations assistant, I would recommend checking the Github issue for any updates or workarounds provided by the Dagster team or the community. If the issue is still open and unresolved, you might consider adding your experience to the discussion or asking for further assistance. For the specific Github issue related to this problem, you can refer to the following link: [Mypy raises error when using

dagster.Config

in Dagster 1.5.5](https://github.com/dagster-io/dagster/issues/17443). This will give you the most up-to-date information on the status of the issue and any potential solutions.