Cli MCP Server

🚀 CLI MCP Server

A secure implementation of the Model Context Protocol (MCP) server for executing commands with comprehensive security features. This server is designed for interaction via the Model Context Protocol.

🚀 Quick Start

The CLI MCP server is designed to provide a secure and feature-rich command execution environment through the Model Context Protocol. It supports features such as command whitelisting, flag validation, and path traversal prevention to ensure commands are run in a controlled environment.

✨ Features

• Command Whitelisting: Allows specifying executable commands and their supported flags.
• Flag Validation: Ensures that the used flags meet expectations.
• Path Traversal Prevention: Prevents malicious operations through relative paths or symbolic links.
• Shell Operator Blocking: Prohibits the use of dangerous shell operators, such as ;, &, etc.
• Command Length Limit: Prevents security risks caused by overly long commands.
• Execution Timeout: Sets the maximum allowed execution time for commands to prevent unresponsive commands.
• Working Directory Restriction: Restricts commands to the specified working directory to prevent out-of-bounds access.

📦 Installation

Dependencies Installation

Use the following command to install project dependencies:

pip install -r requirements.txt

Starting the Server

To start the MCP server, run:

python cli_mcp_server.py

📚 Documentation

Configuration

The server can be configured through the following environment variables:

Available Tools

1. run_command

Used to execute commands in a controlled environment. Supports the following operations:

• Execute allowed commands.
• Validate the validity of flags.
• Ensure path security.

Input Format

{
  "command": "echo",
  "args": ["hello", "world"],
  "flags": ["-n"]
}

2. show_security_rules

Displays the current server's security policy configuration, including allowed commands, flags, and working directory restrictions.

Usage with Claude Desktop

When integrating this server in the Claude desktop environment, configure it according to the following steps:

Development Mode

{
  "mcpServers": {
    "cli-mcp-server": {
      "command": "uv",
      "args": ["run", "cli-mcp-server"],
      "env": {
        "ALLOWED_COMMANDS": "echo,ls,-h",
        "ALLOWED_FLAGS": "-l,-a,--help,--version",
        "MAX_COMMAND_LENGTH": "1024",
        "COMMAND_TIMEOUT": "30"
      }
    }
  }
}

Production Mode

{
  "mcpServers": {
    "cli-mcp-server": {
      "command": "uvx",
      "args": ["cli-mcp-server"],
      "env": {
        "ALLOWED_COMMANDS": "echo,ls,-h",
        "ALLOWED_FLAGS": "-l,-a,--help,--version",
        "MAX_COMMAND_LENGTH": "1024",
        "COMMAND_TIMEOUT": "30"
      }
    }
  }
}

Security Features

• Command Whitelisting: Strictly limits executable commands.
• Flag Validation: Ensures that the used flags meet expectations.
• Path Traversal Prevention: Prevents malicious operations through relative paths or symbolic links.
• Shell Operator Blocking: Prohibits the use of dangerous shell operators, such as ;, &, etc.
• Command Length Limit: Prevents security risks caused by overly long commands.
• Execution Timeout: Sets the maximum allowed execution time for commands to prevent unresponsive commands.
• Working Directory Restriction: Restricts commands to the specified working directory to prevent out-of-bounds access.

Error Handling

The server provides detailed error messages to handle the following situations:

• Security violations (CommandSecurityError)
• Command timeouts (CommandTimeoutError)
• Invalid command formats
• Path security violations
• Execution failures

Error messages will be reported through standard output or environment variables.

This document details how to configure and use the MCP server. For more information, please refer to the source code or contact the development team.