command-executor-mcp-server

command-executor MCP Server

A Model Context Protocol server for executing pre-approved commands securely.

• Secure command execution with pre-approved command list
• Configurable allowed commands through environment variables
• Built with TypeScript and MCP SDK
• Communication via stdio for seamless integration
• Error handling and security validations
• Real-time command output streaming

Install dependencies:

npm install

Build the server:

npm run build

For development with auto-rebuild:

npm run watch

By default, the following commands are allowed:

• git
• ls
• mkdir
• cd
• npm
• npx
• python

You can customize the allowed commands by setting the ALLOWED_COMMANDS environment variable:

export ALLOWED_COMMANDS=git,ls,mkdir,python

To use with Claude Desktop, add the server config:

On MacOS:

~/Library/Application Support/Claude/claude_desktop_config.json

On Windows:

%APPDATA%/Claude/claude_desktop_config.json

Configuration example:

{
  "mcpServers": {
    "command-executor": {
      "command": "/path/to/command-executor/build/index.js"
    }
  }
}

The command-executor server implements several security measures:

1. Pre-approved Command List

   • Only explicitly allowed commands can be executed
   • Default list is restrictive and security-focused
   • Commands are validated by prefix to prevent injection

2. Command Validation

   • Command prefix validation prevents command injection
   • No shell execution for improved security
   • Environment variables are properly sanitized

3. Error Handling

   • Comprehensive error handling for unauthorized commands
   • Clear error messages for debugging
   • Failed commands don't crash the server

4. Environment Isolation

   • Server runs in its own environment
   • Environment variables can be controlled
   • Limited system access

command-executor/
├─ src/
│  └─ index.ts      # Main server implementation
├─ build/
│  └─ index.js      # Compiled JavaScript
├─ assets/
│  └─ header.svg    # Project header image
└─ package.json     # Project configuration

Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector:

npm run inspector

The Inspector will provide a URL to access debugging tools in your browser.

The server provides a single tool:

Executes a pre-approved command.

Parameters:

• command (string, required): The command to execute

Example Request:

{
  "name": "execute_command",
  "arguments": {
    "command": "git status"
  }
}

Example Response:

{
  "content": [
    {
      "type": "text",
      "text": "On branch main\nNothing to commit, working tree clean"
    }
  ]
}

Error Response:

{
  "content": [
    {
      "type": "text",
      "text": "Command execution failed: Command not allowed"
    }
  ],
  "isError": true
}

The server provides detailed error messages for various scenarios:

1. Unauthorized Commands

   {
     "code": "InvalidParams",
     "message": "Command not allowed: [command]. Allowed commands: git, ls, mkdir, cd, npm, npx, python"
   }

2. Execution Failures

   {
     "content": [
       {
         "type": "text",
         "text": "Command execution failed: [error message]"
       }
     ],
     "isError": true
   }

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a new Pull Request

This project is licensed under the MIT License - see the LICENSE file for details.