mcp-registry/mcp-server-commands

    ==================
      
       /// MCP ///
      /// MCP ///
        
    ==================
        
    [server:online]
    [protocol:ready]

mcp-server-commands

by g0t4

Model Context Protocol server that exposes a single run_command tool allowing LLMs to execute arbitrary shell commands via STDIO transport.

181
30
Open Source
GitHubDocumentationNPM

01

run_command

Execute an arbitrary shell command and return its STDOUT and STDERR as text; supports an optional stdin parameter to pass input (e.g., code) or create files.

Installation

1. Prerequisites:
• Node.js >= 18 and npm (or pnpm/yarn)
• Git (to clone the repository)
2. Clone the repository:
git clone https://github.com/g0t4/mcp-server-commands.git cd mcp-server-commands
3. Install dependencies:
npm install # or: pnpm install / yarn install
4. Build TypeScript sources:
npm run build # compiles into ./dist
5. Run the server locally:
npm start # equivalent to: node dist/index.js
6. Configuration (environment variables):
• PORT – HTTP port to expose the MCP endpoint (default 8080)
• AUTH_TOKEN – (optional) bearer token required by clients
• COMMAND_PATH – absolute/relative path of shell commands that the server is allowed to execute.
7. Production deploy:
• Build once (npm run build) and copy the dist/ folder to your host.
• Use a process manager such as PM2:
• Alternatively build a Docker image:
pm2 start dist/index.js --name mcp-server-commands -i max docker build -t mcp-server-commands . docker run -p 8080:8080 -e AUTH_TOKEN=mytoken mcp-server-commands

Documentation

License: MIT License
Updated 7/30/2025