MCP.so
ログイン

MCP Server Tutorial

@cthuaung

MCP Server Tutorial について

概要はまだありません

基本情報

カテゴリ

その他

ランタイム

python

トランスポート

stdio

公開者

cthuaung

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "MCP Server Tutorial": {
      "command": "python",
      "args": [
        "main.py"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is MCP Server Tutorial?

MCP Server Tutorial is a simple MCP (Model Context Protocol) server built with FastMCP that provides various mathematical operations as tools and a dynamic greeting resource. It is intended for developers learning to create and use MCP servers.

How to use MCP Server Tutorial?

Install Python 3.8+, create a virtual environment, and install FastMCP. Run the server with python main.py using SSE transport by default. Connect via MCP Inspector (run npx @modelcontextprotocol/inspector, choose SSE, and connect to http://localhost:8000/sse).

Features of MCP Server Tutorial

  • Basic arithmetic: addition, subtraction, multiplication, division
  • Advanced math: power, square root, cube root, factorial, logarithm, trigonometry
  • Dynamic greeting resource (greeting://{name})
  • Supports SSE and stdio transport protocols
  • Easily extensible with @mcp.tool() and @mcp.resource() decorators

Use cases of MCP Server Tutorial

  • Learning how to build and run an MCP server with FastMCP
  • Prototyping mathematical tool integrations in AI assistants
  • Demonstrating SSE-based MCP connections via Inspector

FAQ from MCP Server Tutorial

What are the prerequisites for running this server?

Python 3.8 or higher and the FastMCP library. Dependencies are installed via uv install fastmcp.

Which transport protocols are supported?

FastMCP supports two transport protocols: stdio (terminal-based) and sse (Server-Sent Events, browser-based compatible with MCP Inspector). The default is SSE.

How can I add new tools or resources?

Add a new tool by using the @mcp.tool() decorator or a new resource by using the @mcp.resource() decorator in main.py.

Where does the server’s data live?

No external databases or storage are mentioned; all operations are stateless calculations performed in memory.

Is authentication or authorization required?

The README does not mention any authentication or authorization mechanisms.

コメント

「その他」の他のコンテンツ