Changed the word client to server in the exercise, and changed the li…

docs/3-AI-Engineering/3.2-mcp.md

@@ -3,8 +3,8 @@ docs/3-AI-Engineering/3.2-mcp.md:
   category: AI Engineering
   estReadingMinutes: 12
   exercises:
-    - name: Create an MCP Client
-      description: Build a simple MCP client that connects to a local MCP server
+    - name: Create an MCP Server
+      description: Build a simple MCP server that connects to a local MCP client
       estMinutes: 45
       technologies:
         - MCP
@@ -125,23 +125,23 @@ Roots provide entry points into your data, helping AI models understand where to
   <div id="chapter-3/3.2/mcp-quiz.js"></div>
 </div>
 
-# Exercise: Create an MCP Client
+# Exercise: Create an MCP Server
 
-In this exercise, you will create a simple MCP client that can connect to a local MCP server. This will help you understand how MCP enables AI applications to interact with external tools and data sources.
+In this exercise, you will create a simple MCP server that can connect to a local MCP client. This will help you understand how MCP enables AI applications to interact with external tools and data sources.
 
 ## Instructions
 
-1. Follow the MCP Client Quickstart tutorial at [https://modelcontextprotocol.io/quickstart/client](https://modelcontextprotocol.io/quickstart/client) to create a Python-based MCP client.
+1. Follow the MCP Server Quickstart tutorial at [https://modelcontextprotocol.io/docs/develop/build-server](https://modelcontextprotocol.io/docs/develop/build-server) to create a Python-based MCP server.
 
 2. The tutorial will guide you through:
    * Setting up your development environment
-   * Creating a basic MCP client that can connect to servers
+   * Creating a basic MCP server that can connect to servers
    * Implementing a query processing system that handles tool calls
    * Building an interactive chat interface
 
-3. Once you have your client working, connect it to one of the sample MCP servers (like Claude Desktop, Windsurf, or VSCode) to test its functionality.
+3. Once you have your server working, connect it to one of the sample MCP clients (like Claude Desktop, Windsurf, or VSCode) to test its functionality.
 
-4. Experiment with modifying the client to handle different types of tools or customize how responses are processed.
+4. Experiment with modifying the server to handle different types of tools or customize how responses are processed.
 
 ## Key Learning Points
 
@@ -152,7 +152,7 @@ In this exercise, you will create a simple MCP client that can connect to a loca
 
 ## Deliverables
 
-* Create a working MCP client based on the tutorial that can connect to at least one MCP server
+* Create a working MCP server based on the tutorial that can connect to at least one MCP client
 * Write a brief explanation of how the MCP client-server architecture enhances AI capabilities
 * Provide one example of how you might extend or customize your MCP client for a specific use case
 * How might the Model Context Protocol change the way developers interact with AI tools in their workflow?

docs/README.md

@@ -375,8 +375,8 @@ docs/3-AI-Engineering/3.2-mcp.md:
   category: AI Engineering
   estReadingMinutes: 12
   exercises:
-    - name: Create an MCP Client
-      description: Build a simple MCP client that connects to a local MCP server
+    - name: Create an MCP Server
+      description: Build a simple MCP server that connects to a local MCP client
       estMinutes: 45
       technologies:
         - MCP