Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is it weird to route all completions to the same function in the high level API? Alternative would be to defer adding the fastmcp decorator for now and do something fancier that routes on URI, but agree that it's nice to bring fastmcp decorator coverage to parity with the lowlevel server.

Also a) maybe it's natural for "real" servers to route all completion requests through the same handler, and b) we could always add new decorators (resource_completion, prompt_completion) later.