mcp-builder

MCP Builder

"MCP turns AI capabilities into composable, discoverable services that any client can use."

When to Use

• Building a new MCP (Model Context Protocol) server
• Defining tools, resources, or prompt templates for AI agents
• Integrating an existing service as an MCP server
• Debugging MCP transport or protocol issues
• Reviewing MCP server implementations for correctness

What is MCP?

• Model Context Protocol is an open standard for connecting AI models to external tools and data
• MCP servers expose capabilities (tools, resources, prompts) over a standardized protocol
• MCP clients (like Claude Desktop, Claude Code, or custom agents) discover and invoke these capabilities
• Transport options: stdio (local), SSE (HTTP streaming), and streamable HTTP

Server Setup (TypeScript SDK)

Installation

npm install @modelcontextprotocol/sdk

Basic Server Structure

import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";

const server = new McpServer({
  name: "my-mcp-server",
  version: "1.0.0",
  capabilities: {
    tools: {},
    resources: {},
    prompts: {},
  },
});

// Register tools, resources, prompts here...

// Connect via stdio transport
const transport = new StdioServerTransport();
await server.connect(transport);

Python SDK Setup

from mcp.server import Server
from mcp.server.stdio import stdio_server

server = Server("my-mcp-server")

# Register handlers with decorators...

async def main():
    async with stdio_server() as (read_stream, write_stream):
        await server.run(read_stream, write_stream)

Tools

What Are Tools?

• Tools are functions that the AI model can invoke to perform actions
• Each tool has a name, description, and input schema (JSON Schema)
• Tools can have side effects (write to DB, call APIs, modify files)

Defining Tools (TypeScript)

import { z } from "zod";

server.tool(
  "get-weather",
  "Get current weather for a city. Use this when the user asks about weather conditions.",
  {
    city: z.string().describe("City name, e.g., 'Istanbul' or 'New York'"),
    units: z.enum(["celsius", "fahrenheit"]).default("celsius")
      .describe("Temperature units"),
  },
  async ({ city, units }) => {
    const weather = await fetchWeather(city, units);
    return {
      content: [
        { type: "text", text: JSON.stringify(weather, null, 2) },
      ],
    };
  }
);