Model Context Protocol (MCP) uses a client-server architecture to connect AI applications (hosts) to various data sources and tools

One to One connections to the MCP Servers

We see in this lecture :

• MCP Features
• How to communicate between the client and server takes place

The MCP Clients that maintain a 1 to 1 connection with MCP servers, The way these communicate with each other is through message defined by MCP itself.

This is live inside our host like Claude desktop or Claude AI, The host is responsible for storing and maintaining all of the clients and connections to the MCP server

Client-Server Architecture

How does it work?

The fundamental piece of protocol starting with tools,

Tools

Tools are very similar with MCP tools. Tools are functions that can be invoked by a clients, These tools allow to retrieving, searching, sending message and update database records. Tools is meant to be data that might require something like a post request or some kind of modification.

Resources

Resource are bit similar to get request (or) Resources are Read-only data or context that exposed by the server. your application to choose weather can consume or use these resources. but, It doesn’t necessary necessarily have to bring it into context. Example Resource : Data Base, API Response, Files, PDF’s and so on.