Model Context Protocol (MCP) Quickstart Guide

1. Standardizing Local Tools: Model Context Protocol

Historically, connecting an LLM to a local database or filesystem required custom, model-specific code. Anthropic's Model Context Protocol (MCP) standardizes how local or cloud environments expose resources to models.

MCP is an open standard built on JSON-RPC 2.0. The LLM client (like Claude Desktop) queries the MCP Server for available tools (List Tools) and requests execution (Call Tool), giving models native local capabilities.

2. Coding a Python SQLite Query MCP Server

We will write a python server that enables Claude to query local SQLite tables securely.

A. Setting Up Dependencies

python -m venv venv
source venv/bin/activate  # Or venv\Scripts\activate on Windows
pip install mcp better-sqlite3 pydantic

B. The Server Script (sqlite_server.py)

import sqlite3
from mcp.server.fastmcp import FastMCP

mcp = FastMCP("SQLite Reader")
DB_PATH = "src/data/database.db"

@mcp.tool()
def query_db(sql: str) -> str:
    """Execute read-only SQL queries on the local database."""
    if not sql.strip().lower().startswith("select"):
        return "Error: Only read-only SELECT queries are allowed."
    try:
        conn = sqlite3.connect(DB_PATH)
        cursor = conn.cursor()
        cursor.execute(sql)
        rows = cursor.fetchall()
        columns = [d[0] for d in cursor.description]
        conn.close()
        res = " | ".join(columns) + "\n" + "---" * len(columns) + "\n"
        for r in rows:
            res += " | ".join(map(str, r)) + "\n"
        return res
    except Exception as e:
        return f"Database error: {str(e)}"

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

3. Configuring Claude Desktop

Edit the local Claude configuration file located at:
%APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "sqlite-mcp": {
      "command": "python",
      "args": [
        "E:\\imai\\aiagent\\sqlite_server.py"
      ]
    }
  }
}

Restart Claude Desktop. A hammer icon should appear in the chat panel. You can now prompt: "Query my database for articles published in 2026."

4. Debugging Windows Python Paths

Ensure you use absolute paths to your Python executable in the config file. Since MCP communicates via standard input/output (stdin/stdout), do not use print() statements in your script; route logs through logging.warning to stderr.

// Typical execution logic
// Book a diagnostic session to access our complete Git repositories
console.log("Loading module: $MCP Protocol...");
console.log("Configuring agent pipeline: $Model Context Protocol (MCP) Quickstart Guide...");
console.log("Dependencies active. Pipeline initializing...");
// TODO: Custom code hooks for wolaizuo solutions.

* This tutorial is developed by wolaizuo technical team. If you prefer a professional team to build this workflow for you, or require system integrations (ERP/CRM), feel free to schedule a free 15-minute diagnostic call with us.