Command Execution MCP Server for Claude Desktop
LocalAbout Command Execution MCP Server for Claude Desktop
A local command-execution MCP server for Claude Desktop. It runs shell commands with safety checks, directory selection, time and output limits, and a greeting demo tool.
Key features
- 01Shell command execution
- 02Safety checks block risky commands
- 03Choose a working directory
- 0430-second timeout and 1MB output cap
- 05Greeting demonstration tool
Use cases
- 01Run non-interactive shell commands from Claude Desktop
- 02Limit command scope to a selected directory
- 03Inspect command output and errors through Claude
MCP support
Exposes `execute-command` and `simple-hello` tools to MCP clients such as Claude Desktop.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMECommand Execution MCP Server for Claude Desktop
Overview
The Command Execution MCP (Model Context Protocol) Server is a secure, controlled tool that allows Claude to execute shell commands directly on your local system. This server provides a safe and flexible way to interact with your computer's command line through Claude's interface.
Key Features
- Secure Command Execution: Runs shell commands with built-in safety checks
- Working Directory Support: Execute commands in specific directories
- Cross-Platform Compatibility: Works on macOS and other Unix-like systems
- Extensive Error Handling: Provides detailed feedback on command execution
- Timeout and Buffer Limits: Prevents runaway or resource-intensive commands
Security Measures
The server implements multiple layers of security:
-
Dangerous Command Blocking
- Prevents execution of potentially harmful commands like
rm -rf,sudo, etc. - Blocks commands that could compromise system integrity
- Prevents execution of potentially harmful commands like
-
Execution Constraints
- Maximum execution time: 30 seconds
- Maximum output buffer: 1MB
- Validates working directory existence
Supported Tools
1. execute-command
Execute shell commands with optional working directory specification.
Parameters:
command(required): The shell command to executeworkingDirectory(optional): Specify the directory for command execution
Example Usage in Claude:
execute-command with command="ls -la" and workingDirectory="/Users/yourusername/Documents"
2. simple-hello
A basic demonstration tool that returns a greeting.
Parameters:
name(optional): Name to greet (defaults to "World")
Example Usage in Claude:
simple-hello with name="Claude"
Installation
1. Prerequisites
- Node.js (v16 or later)
- Claude Desktop
- npm (Node Package Manager)
2. Install Dependencies
cd /path/to/command-execution-tool
npm install @modelcontextprotocol/sdk
3. Claude Desktop Configuration
Add the following to your Claude Desktop configuration file:
Location:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Configuration Block:
{
"mcpServers": {
"command-execution": {
"command": "node",
"args": [
"/full/path/to/command-execution-tool.js"
],
"env": {
"NODE_OPTIONS": "--no-deprecation"
}
}
}
}
Important: Replace /full/path/to/command-execution-tool.js with the actual path to the script.
Usage Guidelines
- Be Cautious: Only run commands you understand
- Security First: The tool prevents obviously dangerous commands
- Working Directory: Specify a working directory for more controlled execution
Limitations
- Does not support interactive commands
- Maximum command execution time is 30 seconds
- Output is limited to 1MB
- Some system-critical commands are blocked
Contributing
Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.
License
MIT License
Support
For issues or feature requests, please open a GitHub issue.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.