MCP Servers Explained: How AI Assistants Connect to Your Tools (With Examples)

#ai #tutorial #webdev #python

If you've used Claude, Cursor, or Windsurf recently, you've probably seen "MCP" mentioned. But most explanations are either too abstract or too deep in the spec.

Here's the practical version.

What Is MCP?

Model Context Protocol is a standard that lets AI assistants use external tools. Think of it as USB for AI:

Before MCP:  AI → custom integration → Tool A
             AI → different integration → Tool B
             AI → yet another → Tool C

With MCP:    AI → MCP → Any Tool

One protocol, infinite tools.

Why Should You Care?

If you build tools, APIs, or services — MCP lets AI assistants use them directly. Your users can say "search my database" or "create a ticket" and the AI handles it.

If you use AI assistants — MCP gives them superpowers. Instead of copy-pasting between tools, the AI does it for you.

A Simple MCP Server in Python

from mcp.server import Server
from mcp.types import Tool, TextContent
import requests

server = Server("weather-server")

@server.tool()
async def get_weather(city: str) -> list[TextContent]:
    """Get current weather for any city."""
    # Using Open-Meteo (free, no API key)
    geo = requests.get(
        "https://geocoding-api.open-meteo.com/v1/search",
        params={"name": city, "count": 1}
    ).json()

    if not geo.get("results"):
        return [TextContent(type="text", text=f"City '{city}' not found")]

    lat = geo["results"][0]["latitude"]
    lon = geo["results"][0]["longitude"]

    weather = requests.get(
        "https://api.open-meteo.com/v1/forecast",
        params={"latitude": lat, "longitude": lon, "current_weather": "true"}
    ).json()["current_weather"]

    return [TextContent(
        type="text",
        text=f"{city}: {weather['temperature']}C, wind {weather['windspeed']} km/h"
    )]

@server.tool()
async def get_crypto_price(coin: str) -> list[TextContent]:
    """Get current cryptocurrency price."""
    data = requests.get(
        "https://api.coingecko.com/api/v3/simple/price",
        params={"ids": coin.lower(), "vs_currencies": "usd"}
    ).json()

    price = data.get(coin.lower(), {}).get("usd", "not found")
    return [TextContent(type="text", text=f"{coin}: ${price}")]

if __name__ == "__main__":
    server.run()

How to Use It

Add to your Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "weather": {
      "command": "python",
      "args": ["weather_server.py"]
    }
  }
}

Now Claude can check weather and crypto prices directly.

Real MCP Servers You Can Use Today

Server	What It Does
GitHub MCP	Manage repos, issues, PRs
Filesystem	Read/write local files
PostgreSQL	Query databases
Notion	Manage pages and databases
Slack	Send and read messages
Brave Search	Web search
Playwright	Browser automation