server-run-commands
About
server-run-commands is an MCP (Model Context Protocol) server that lets an LLM execute commands on the local operating system. It is designed for developers who want to give a language model the ability to run shell commands and receive their output.
Details
- Author
- anton-107
- Downloads
- 421
- Categories
- Other
Jump to
- Runs commands on the local OS
- Passes the process exit code back to the LLM
- Returns command stdout to the LLM
- Simple installation via npm and build step
Clone the repository, then run npm install and npm run build to compile the server. To use it with Claude Desktop, add an entry to claude_desktop_config.json specifying the path to Node.js and the build folder. The server exposes a single tool called run-command that accepts a command (string) input.
server-run-commands
MCP server to run commands on the local OSCode written following the official MCP guide:
- https://modelcontextprotocol.io/quickstart
Tools
- run-command - Run a command on the local OS - Input:command (string)
- Passes the process exit code and stdout back to the LLM
Installation
-git clone https://github.com/anton-107/server-run-commands.git
- cd server-run-commands
- npm install
- npm run build
Usage with Claude Desktop
Add this to yourclaude_desktop_config.json:
```json
{
"mcpServers": {
"run-commands": {
"command": "<PATH TO LOCAL NODE>",
"args": [
"<PATH TO GIT CLONE FOLDER>/server-run-commands/build"
]
}
}
}Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
