MCP server to run arbitrary commands on the host machine via a single tool.
mcp-server-commands is easy to set up with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view g0t4/mcp-server-commands --webOpen the official repository or website.
Check the README for package manager, auth, and platform requirements.
Try it in a small test task inside your agent workflow.
This tool lets AI assistants run commands on your computer, like listing files or running scripts. You can either type a command as you would in a terminal, or provide the program and arguments separately. It returns the output so the AI can use it.
mcp-server-commands is a Model Context Protocol (MCP) server that exposes a single tool called `runProcess` for executing commands on the host machine. It supports two mutually exclusive invocation methods: `command_line` (a string executed via the system's default shell, supporting pipes, redirects, and variable expansion) and `argv` (an array where the first element is the executable and the rest are arguments, bypassing shell interpretation). The tool returns both STDOUT and STDERR as text, and optionally accepts STDIN input, enabling LLMs to pass scripts to interpreters like bash, python, or fish. It is designed for use with MCP-compatible clients such as Claude Desktop, Groq Desktop, and Zed. The server is written in TypeScript and published on npm. Security warnings emphasize that users should review commands before approval and avoid running with sudo.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 26 days ago.
226 GitHub stars indicate community interest.
7 open issues signal maintenance load.
MIT license detected.
Run system commands like `hostname` or `ls -al` from an AI chat interface.
Execute scripts by piping them via STDIN to interpreters like bash or python.
Create files using commands like `cat >> file.txt` with content provided by the AI.
Automate development tasks such as building projects or running tests.
Retrieve command outputs to inform AI responses or debugging.
Arbitrary command execution can lead to data loss or system compromise if misused.
Always review commands before approving; do not run with sudo.
The server runs with the permissions of the user that starts it.
226
Stars
34
Forks
7
Issues
MIT
License
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
A secure MCP server for AI applications to interact with MySQL databases.
Up-to-date code documentation for LLMs and AI code editors, eliminating outdated or hallucinated API references.
3 security/trust notes recorded.
Setup difficulty is 2/5.