In this article, we'll explore the integration of Large Language Models and systems built on top of them. The key concept in this space recently is the Model Context Protocol (MCP).
(…) Different servers should be able to work together. Servers should have limited access to user conversations and shouldn’t access other servers. (…)
I wasn’t sure what to make of that either. I just took it as the idea that you can have servers acting on different parts of the conversation without interfering with each other, but I’m not sure how to enforce that. If you have two different “branding” servers trying to change colors or styling, how do they interact without accessing each others output? Is there a quarterback needed?
i can’t grasp this, looks contradictory to me :
I wasn’t sure what to make of that either. I just took it as the idea that you can have servers acting on different parts of the conversation without interfering with each other, but I’m not sure how to enforce that. If you have two different “branding” servers trying to change colors or styling, how do they interact without accessing each others output? Is there a quarterback needed?