There is a difference between protocol error and tool usage error, makes sense you want the model to see the tool usage error, so they can correct.
I'm guessing it has a the same shape as a normal message + IsError so on the handling side you don't really have to do anything special to handle it, just proceed as normal and send the results to the LLM so it can correct if needed.
Because they don't have exclusive control, unlike social media where you can't take your data and move it to another provider, you can just take your repo to whichever provider or self-hosted GitOps option you want.
Probably architecture changes (x86 has a lot of historic baggage that difficults newer designs) and also more specialized hardware in the CPU, probably this might also be one of the reasons Apple went this way with the M Silicon
https://modelcontextprotocol.io/specification/2025-06-18/ser...