Allows executing Git commands such as checking repository status and managing version control directly through the MCP server.
Enables starting, monitoring, and managing Node.js applications and development servers asynchronously.
Supports running npm commands for package management and executing project scripts synchronously or asynchronously.
Provides capabilities to start and monitor Python-based development servers and scripts.
Allows for the execution and monitoring of long-running Webpack build processes.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Run Command MCP Serverrun npm install and show me the output"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Run Command MCP Server
A Model Context Protocol (MCP) server that provides command execution capabilities with both synchronous and asynchronous streaming support.
Features
Synchronous Command Execution: Run commands and wait for completion
Asynchronous Command Execution: Start commands in background and monitor progress
Real-time Output Streaming: Check command output while it's running
Process Management: List, monitor, kill, and clean up processes
Timeout Support: Automatically terminate long-running commands
Multiple Transport Modes:
Stdio Transport: Standard MCP communication via stdin/stdout
HTTP Transport: RESTful API with Server-Sent Events (SSE) support for real-time notifications
MCP Standard Compliance: Fully compliant with MCP specifications
Usage
Stdio Transport (Default)
Claude Desktop
Or use locally after cloning:
Github Copilot
Or use locally after cloning:
HTTP Transport (Streamable)
Start the HTTP server:
The server will be available at: http://localhost:8123/mcp
Github Copilot Configuration (HTTP)
Available Tools
run_command
Run a custom shell command synchronously and return the output (stdout, stderr, exit code). Blocks until command completes.
Parameters:
command(required): The full command to execute (e.g., "echo hello world", "npm install", "git status")timeout(optional): Timeout in milliseconds (default: 30000)
Example:
start_command
Start a command asynchronously (non-blocking). Returns a process_id to check status and output later.
Parameters:
command(required): The full command to execute (e.g., "node server.js", "npm run dev")timeout(optional): Timeout in milliseconds. 0 means no timeout (default: 0)
Example:
get_command_output
Get the current output and status of a running or completed async command by process_id.
Parameters:
process_id(required): The process ID returned by start_commandtail(optional): Only return the last N lines of output (0 = all)
Example:
list_processes
List all tracked processes (running, completed, failed, etc.)
Parameters:
status(optional): Filter by status: running, completed, failed, killed, error, timed_out
Example:
kill_process
Kill a running process by process_id
Parameters:
process_id(required): The process ID to kill
Example:
clear_processes
Clear finished processes from memory. If process_id is provided, clears that specific process. Otherwise clears all non-running processes.
Parameters:
process_id(optional): Specific process ID to clear
Example:
Usage Examples
Example 1: Run a simple synchronous command
Example 2: Start a long-running server
Example 3: Monitor build progress
Example 4: Run Azure Functions locally
Use Cases
Development Servers: Start and monitor Node.js, Python, or other development servers asynchronously
Build Processes: Run and monitor long build processes (webpack, tsc, etc.)
Testing: Run test suites and monitor results
System Administration: Execute system commands and check results
Log Monitoring: Start services and continuously check their logs