PluginBench
MCP Server
Active

Remote Desktop Commander MCP Server

app.desktopcommander/remote-desktop-commander

Control your computer's files and terminal from Claude, ChatGPT, and other AI apps via a secure remote MCP server.

What is the Remote Desktop Commander MCP server?

Remote Desktop Commander is a hosted MCP server that gives AI assistants like Claude and ChatGPT secure access to your computer's filesystem and terminal from anywhere. It relays commands through OAuth 2.0 authentication to a device agent running on your machine, enabling file management, code execution, and system automation without local setup.

Remote Desktop Commander bridges your AI assistant to your personal computer through a hosted MCP endpoint. You run a lightweight device agent on your machine, then connect any AI app (Claude, ChatGPT, Cursor, VS Code) to the remote server. The AI can then read/write files, search directories, run terminal commands, manage processes, and control multiple paired machines—all with your user permissions and only while the agent is running.

How to install Remote Desktop Commander

Copy-paste configuration for popular MCP clients.

transport: http
Config generated by PluginBench — verify against the source before use.
~/.cursor/mcp.json
{
  "mcpServers": {
    "remote-desktop-commander": {
      "url": "https://mcp.desktopcommander.app/mcp"
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • Read File — Read the contents of a file on your computer
  • Read Multiple Files — Read contents of multiple files at once
  • List Directory — List the contents of a directory
  • Get File Info — Get metadata and information about a file
  • Start Search — Start a search for files in your filesystem
  • Get More Search Results — Retrieve additional search results
  • Stop Search — Stop an active search
  • List Active Searches — List all currently running searches
  • Write File — Write or create a file on your computer
  • Edit File Block — Edit a specific block or section of a file
  • Move or Rename File — Move or rename a file or directory
  • Create Directory — Create a new directory
  • Write or Modify PDF — Write or modify PDF files
  • Start Process — Start a new process or application
  • Send Input to Process — Send input to a running process
  • Read Process Output — Read output from a running process
  • List Active Sessions — List all active terminal sessions
  • List System Processes — List running system processes
  • Terminate Session — Terminate an active terminal session
  • Kill System Process — Kill a running system process

Use cases

  • Create and run scripts on your computer from an AI chat, automating development tasks and system administration
  • Explore and understand large codebases by having the AI navigate your filesystem and read files in context
  • Edit documents in place (Word, Excel, PDF) stored on your computer without downloading or uploading
  • Manage multiple development machines by pairing several computers and controlling them by name in one conversation
  • Search your file system and have the AI analyze and organize files modified in specific timeframes

Remote Desktop Commander MCP server FAQ

What is Remote Desktop Commander?

Remote Desktop Commander is a hosted MCP server that connects AI assistants (Claude, ChatGPT, Cursor) to your personal computer. It relays file and terminal commands securely through OAuth 2.0, giving your AI full access to your filesystem and shell while you're away from the keyboard.

Is Remote Desktop Commander free?

The README does not specify pricing. The service is currently in beta; check mcp.desktopcommander.app for current pricing and availability.

How do I install it in Cursor or Claude?

In Cursor, click the 'Add to Cursor' badge in the README, or manually add the URL `https://mcp.desktopcommander.app/mcp` to `~/.cursor/mcp.json`. In Claude Desktop, go to Settings → Connectors → Add custom connector and enter the same URL. Both require OAuth sign-in on first use.

What authentication is required?

Remote Desktop Commander uses OAuth 2.0 with PKCE. On first setup, you run `npx @wonderwhy-er/desktop-commander@latest remote` on your computer, verify a pairing code in your browser, and then sign in via OAuth when connecting from your AI app.

Does my computer need to be running all the time?

The device agent must be running in a terminal on your computer for the AI to access it. Stop the agent with Ctrl+C and access ends immediately. You can review and revoke paired devices anytime from the dashboard.

What operating systems are supported?

Remote Desktop Commander works on macOS, Windows, and Linux. It requires Node.js 18 or newer to run the device agent via `npx`.

README (reference)

Source of truth, from the repository.

Remote Desktop Commander

Remote Desktop Commander

Build, explore, automate on your computer — from anywhere.

Reach your computer's filesystem and terminal from any AI app that supports remote MCP connectors. Remote Desktop Commander is a hosted MCP server that relays a secure connection to your real machine, giving your AI assistant the same full file and shell access as the local server — without you needing to be at the keyboard.

Add to Cursor

This repository hosts the public manifests, documentation and issue tracker for the hosted service. The service implementation is not open source. For the open-source local (stdio) server, see DesktopCommanderMCP.

What you can do

  • Build from scratch — create features and applications with simple commands
  • Manage development environments — set up servers, configure systems, handle processes
  • Explore existing codebases — navigate and understand complex repositories
  • Edit documents in place — work on Word, Excel, and PDF files where they live on disk
  • Manage context and documentation — keep track of project details and technical specifications
  • Control multiple machines — pair several computers and address them by name in one conversation

Quick start

1. Start the device agent

On the computer you want to control:

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

Your browser opens to a verification page showing a pairing code (also printed in the terminal — a standard OAuth device-flow check). Sign in, confirm the codes match, and the terminal reports "Device ready". Keep the agent running: close the terminal, no access.

2. Connect your AI client

Each client walks you through OAuth sign-in on first connect.

claude.ai / Claude Desktop — Settings → Connectors → Add custom connector, then enter:

https://mcp.desktopcommander.app/mcp

Claude Code:

claude mcp add --transport http remote-desktop-commander https://mcp.desktopcommander.app/mcp

(If you work inside a checkout of this repo, Claude Code also picks up the included .mcp.json automatically.)

ChatGPT — Settings → Apps & Connectors → Advanced settings → enable Developer mode (requires a plan that includes it), then Create connector and enter:

https://mcp.desktopcommander.app/mcp

Cursor — one click: Add to Cursor. Or add to ~/.cursor/mcp.json manually:

{
  "mcpServers": {
    "remote-desktop-commander": {
      "url": "https://mcp.desktopcommander.app/mcp"
    }
  }
}

VS Code — install the agent plugin (bundles this server):

copilot plugin install desktop-commander/remote-desktop-commander

Or add the server directly to .vscode/mcp.json in your workspace (or your user-level mcp.json):

{
  "servers": {
    "remote-desktop-commander": {
      "type": "http",
      "url": "https://mcp.desktopcommander.app/mcp"
    }
  }
}

Gemini CLI:

gemini extensions install https://github.com/desktop-commander/remote-desktop-commander

3. Try it

These work against any machine running the agent, no setup needed:

"List the contents of my home directory and tell me what's there."

"Create a folder called ai-test in my Documents folder. Inside it, write a short Python script that prints the date, and run it."

"Start a Python REPL on my machine, import pandas, and tell me which version is installed. Keep the session alive."

"Search my home directory for files modified in the last 24 hours and group them by folder."

Full walkthrough with troubleshooting: docs/SETUP.md

Requirements

  • Node.js 18 or newer (for npx) on each computer you want to control
  • One terminal running the device agent per machine
  • Works on macOS, Windows, and Linux

Available tools

CategoryTools
Read & exploreRead File, Read Multiple Files, List Directory, Get File Info
SearchStart Search, Get More Search Results, Stop Search, List Active Searches
Write & editWrite File, Edit File Block, Move or Rename File, Create Directory, Write or Modify PDF
Terminal & processesStart Process, Send Input to Process, Read Process Output, List Active Sessions, List System Processes, Terminate Session, Kill System Process
Devices & accountList Paired Devices, Current User Info, Ping Device, Shutdown Device Agent
Configuration & helpGet Configuration, Set Configuration Value, Get Prompt Library, Get Recent Tool Activity, Get Usage Statistics, Submit Feedback

How it works

sequenceDiagram
    participant AI as AI app (web, mobile, desktop)
    participant S as mcp.desktopcommander.app
    participant D as Your computer (device agent)
    AI->>S: MCP tool call (OAuth 2.0 bearer)
    S->>D: relayed to your paired device
    D->>D: executes locally
    D->>S: result
    S->>AI: MCP response
  • Authentication is standard OAuth 2.0 with PKCE; devices pair via the OAuth device authorization flow (the short-code approval you see at setup).
  • The server only accesses data explicitly requested through tool calls; transmission is encrypted in transit (HTTPS/TLS).
  • A machine is reachable only while its device agent is running — stop it with Ctrl+C and access ends. Devices can also be reviewed and revoked at any time from the dashboard.
  • Tools execute with your user account's permissions on the paired machine, so treat your connected AI accounts like credentials and protect them with MFA. Read the full trust model in SECURITY.md.

Support & legal

Related

Related MCP servers

DEdevglow logo

devglow

Active

Start, stop, and inspect local dev servers in a desktop app instead of the agent's session.

0
View repository →
NOnoglow logo

noglow

Active

Drop a short text fragment into a scratchpad slot, with newlines intact.

0
View repository →
TOtodoglow logo

todoglow

Active

See what each of your coding agents is working on, and read or advance your todos.

0
View repository →

Query and troubleshoot your local network in plain language: 37 tools, strictly local, no cloud.

Read-only MCP over your own revenue and net margin across 14 billing rails

Experimental. Plan scuba decompression dives, gas blends and gas use with Dive Kit. Reference only.