home / mcp / claude mcp command server

Claude Mcp Command Server

Python mcp server

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "kimjungyeol-mcp_command_server": {
      "command": "poetry",
      "args": [
        "run",
        "python",
        "src/mcp_command_server/mcp_server_api.py"
      ]
    }
  }
}

The Claude MCP Command Server lets you run a local command-execution endpoint that you can talk to from an MCP client. It executes your commands through Poetry-managed Python code, making it easy to test and integrate command-oriented workflows in your MCP setup.

How to use

To operate this MCP server, run the Python API that exposes the command execution endpoint. Your MCP client can request commands to be executed, and the server will run them in a controlled environment and return results. Use it to automate local command workflows, test CLI interactions, or integrate command execution into larger MCP client pipelines.

How to install

Prerequisites - Python is installed on your machine - Poetry is installed for dependency management (see commands below)

# 의존성 설치
poetry install

# MCP 패키지가 PyPI에 없는 경우
poetry add git+https://github.com/modelcontextprotocol/python-sdk.git

가상환경 활성화 및 실행 테스트

# Poetry 쉘 활성화
poetry shell

# 또는 명령어 실행 테스트
poetry run python src/mcp_command_server/mcp_server_api.py

Additional sections

Configuration and runtime options for the MCP Command Server are provided through stdio-based execution commands. You can run the server directly via Poetry, or use a Windows command to navigate to the project directory and start the server from there.

{
  "mcpServers": {
    "command_executor": {
      "type": "stdio",
      "command": "poetry",
      "args": ["run", "python", "src/mcp_command_server/mcp_server_api.py"]
    },
    "command_executor_win": {
      "type": "stdio",
      "command": "cmd",
      "args": [
        "/c",
        "cd /d C:\\Users\\[user]\\python\\mcp_command_server && poetry run python src/mcp_command_server/mcp_server_api.py"
      ]
    }
  }
}

Notes and tips

The server runs the MCP command API locally. Use the standard Poetry workflow to manage dependencies, and ensure your working directory contains the mcp_command_server project when starting the server with the Windows command.

Available tools

install_dependencies

Install dependencies with Poetry, adding the MCP SDK if it's not available on PyPI.

start_server

Start the MCP Command Server using the designated Python entry point via Poetry to run the API.