Desktop Commander MCP 🔌 MCP server Open source

What it does

Desktop Commander is an MCP server that gives an AI assistant hands-on control of your local machine: a terminal, a filesystem, and surgical code editing, rather than just a chat window. It was built with Claude Desktop in mind but speaks plain MCP, so any MCP-capable client can use it.

Tools it gives your agent

• Terminal execution with interactive process control (start, monitor and manage long-running commands)
• File read, write and search across the filesystem
• Surgical code edits using search/replace diff blocks instead of full-file rewrites
• In-memory code execution for Python, Node.js and R
• Process management for long-running tasks
• Native reading of Excel, PDF and DOCX files

Works with

The project documents setup for Claude Desktop, Claude web and ChatGPT (both via remote MCP), Cursor, Windsurf, VS Code/GitHub Copilot, Cline, JetBrains IDEs, and other MCP-compatible clients generally.

How to install or connect

npx @wonderwhy-er/desktop-commander@latest setup

macOS/Linux script install:

curl -fsSL https://raw.githubusercontent.com/wonderwhy-er/DesktopCommanderMCP/refs/heads/main/install.sh | bash

Docker:

bash <(curl -fsSL https://raw.githubusercontent.com/wonderwhy-er/DesktopCommanderMCP/refs/heads/main/install-docker.sh)

Maintenance and safety

Active project (9,600+ stars, weekly commits, MIT licensed). This server grants an AI agent real terminal and filesystem access on your machine, which is powerful but should be scoped and reviewed carefully, especially before pointing it at a machine with sensitive data; prefer running it in a container or a disposable environment for anything you don't fully trust the agent with.

Who should use it

Developers who want their AI assistant to actually run commands, edit files and manage local processes directly, instead of copy-pasting code back and forth between a chat window and a terminal.