Overview of FastMCP Components: Tools vs. Resources vs. Prompts

📘Introduction

Welcome to this hands-on FastMCP tutorial! In this guide, we'll walk you step-by-step through FastMCP's core components and how they interoperate to build powerful MCP servers. By the end, you’ll understand what each component does and how to use them effectively in a Python with FastMCP.

✅ Prerequisites

Before you begin, make sure you have the following:

🐍☑️ Installed Python
📦☑️ Installed Pip
🌐☑️ Created and activated a virtual environment (venv)
⚡☑️ FastMCP installed inside your venv
⚙️☑️ Created a FastMCP Server instance

💡What are components in FastMCP?

FastMCP implements the Model Context Protocol (MCP), providing a Pythonic and standardized way to expose "what your MCP server can do" to clients and LLM-enabled agents—including tools, data resources, and message templates.

The three components available in FastMCP are:

🛠️ Tools - executable actions (e.g., functions or APIs) the model can call
📂 Resources - contextual or static/dynamic data exposed to the model
🗯️ Prompts - reusable, parameterized message templates to guide interactions

Let's have a closer look at each component and how you can use them in Python.

🛠️ Tools

Tools are executable functions you expose via FastMCP that LLMs can discover and invoke—for instance, performing calculations, fetching external data, or performing API calls

You can simply add tools to your FastMCP Server using the mcp.tool decorator. Here is a simple example: