Mcp_server

Name: Mcp_server
Author: hfhfn

@hfhfn

リポジトリ

Mcp_server について

快速上手MCP

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

hfhfn

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-hfhfn": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp[cli]",
        "mcp",
        "run",
        "path/to/mcp.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?

Mcp_server is a MCP (Model Context Protocol) server configuration reference that demonstrates how to run MCP services using uv/uvx and connect clients via stdio, SSE, or streamable HTTP transports. It provides example configurations for multiple MCP tools such as math, weather, 12306-mcp, sequentialthinking, quickchart-server, and excel-mcp-server.

How to use Mcp_server?

Run the MCP service with uv run --with mcp[cli] mcp run path/to/mcp.py or uvx --with pillow --with mcp[cli] fastmcp run path/to/mcp.py. The service listens on port 8000 by default, offering SSE at /sse and streamable HTTP at /mcp. Configure your MCP client with one of the three transport modes (stdio, SSE, streamable HTTP) and the appropriate URL or command/args.

Key features of Mcp_server

Supports uv/uvx for Python environment management
Offers SSE and streamable HTTP URL modes
Provides configuration examples for multiple MCP tools
Supports stdio, SSE, and streamable HTTP transports
Includes environment variable configuration for excel-mcp-server
Uses Bearer token authentication for remote MCP services

Use cases of Mcp_server

Run a local MCP service and connect a client via stdio
Access the service from a Docker container using host.docker.internal
Configure multiple MCP tools (math, weather, 12306, sequential thinking, charting, Excel) in one client
Set environment variables like cache, log path, and retention for an Excel MCP server
Connect to remote MCP services using SSE with Bearer token authentication

FAQ from Mcp_server

How do I start the MCP service?

Use uv run --with mcp[cli] mcp run path/to/mcp.py or uvx --with pillow --with mcp[cli] fastmcp run path/to/mcp.py.

What transport modes are supported?

The service supports stdio, SSE (Server-Sent Events), and streamable HTTP modes.

How do I configure a stdio‑based MCP client?

Use uv as the command with --directory pointing to the project root and run main.py as arguments.

Can I set environment variables for MCP tools?

Yes. The excel-mcp-server example shows variables such as LOG_PATH, CACHE_MAX_AGE, CACHE_CLEANUP_INTERVAL, LOG_RETENTION_DAYS, and LOG_CLEANUP_INTERVAL.