Author: Furkan Karacan - https://www.linkedin.com/in/furkan-karacan/

What we’ll be building

Purpose: In this tutorial, we’ll create a complete MCP (Model Context Protocol) server from scratch that demonstrates how LLMs can interact with external APIs through standardized tools.

What it achieves: By the end of this guide, you’ll have a working MCP server that:

• Connects to the Random User API to fetch fake user data
• Exposes a tool called get_random_user_data that LLMs can use
• Integrates seamlessly with VS Code as an MCP host
• Follows all MCP best practices for production-ready code

The end result: We’ll build a server that exposes the tool get_random_user_data, then connect it to VS Code where you can interact with it through an AI assistant.

Note: When you are stuck, you can use Step-by-step Screenshots page for detailed visual guidance on registering and testing your MCP server in VS Code.

What is MCP?

Core MCP Concepts

Purpose: Understanding the fundamental building blocks of MCP will help you design and build effective servers.

What it achieves: This section establishes the conceptual foundation you need to understand how MCP servers work and what they can do.

MCP servers can provide three main types of capabilities: