{
  "markdown": "# Remote Desktop Commander\n\n![Remote Desktop Commander](assets/logo-small.png)\n\n**Build, explore, automate on your computer — from anywhere.**\n\nReach your computer's filesystem and terminal from any AI app that supports remote MCP connectors. Remote Desktop Commander is a hosted [MCP](https://modelcontextprotocol.io) 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.\n\n- **Server endpoint:** `https://mcp.desktopcommander.app/mcp` (Streamable HTTP, OAuth 2.0)\n- **Dashboard:** [mcp.desktopcommander.app](https://mcp.desktopcommander.app)\n- **Status:** currently in beta\n- **Demo:** [watch the video](https://youtu.be/C2kH5itFsGg)\n\n[![Add to Cursor](https://img.shields.io/badge/Add_to-Cursor-000000?style=for-the-badge&logo=cursor&logoColor=white)](cursor://anysphere.cursor-deeplink/mcp/install?name=remote-desktop-commander&config=eyJ1cmwiOiAiaHR0cHM6Ly9tY3AuZGVza3RvcGNvbW1hbmRlci5hcHAvbWNwIn0=)\n\n> 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](https://github.com/wonderwhy-er/DesktopCommanderMCP).\n\n## What you can do\n\n- **Build from scratch** — create features and applications with simple commands\n- **Manage development environments** — set up servers, configure systems, handle processes\n- **Explore existing codebases** — navigate and understand complex repositories\n- **Edit documents in place** — work on Word, Excel, and PDF files where they live on disk\n- **Manage context and documentation** — keep track of project details and technical specifications\n- **Control multiple machines** — pair several computers and address them by name in one conversation\n\n## Quick start\n\n### 1. Start the device agent\n\nOn the computer you want to control:\n\n```bash\nnpx @wonderwhy-er/desktop-commander@latest remote\n```\n\nYour 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.\n\n### 2. Connect your AI client\n\nEach client walks you through OAuth sign-in on first connect.\n\n**claude.ai / Claude Desktop** — Settings → Connectors → **Add custom connector**, then enter:\n\n```\nhttps://mcp.desktopcommander.app/mcp\n```\n\n**Claude Code:**\n\n```bash\nclaude mcp add --transport http remote-desktop-commander https://mcp.desktopcommander.app/mcp\n```\n\n(If you work inside a checkout of this repo, Claude Code also picks up the included [`.mcp.json`](.mcp.json) automatically.)\n\n**ChatGPT** — Settings → Apps & Connectors → Advanced settings → enable **Developer mode** (requires a plan that includes it), then **Create connector** and enter:\n\n```\nhttps://mcp.desktopcommander.app/mcp\n```\n\n**Cursor** — one click: [**Add to Cursor**](cursor://anysphere.cursor-deeplink/mcp/install?name=remote-desktop-commander&config=eyJ1cmwiOiAiaHR0cHM6Ly9tY3AuZGVza3RvcGNvbW1hbmRlci5hcHAvbWNwIn0=). Or add to `~/.cursor/mcp.json` manually:\n\n```json\n{\n  \"mcpServers\": {\n    \"remote-desktop-commander\": {\n      \"url\": \"https://mcp.desktopcommander.app/mcp\"\n    }\n  }\n}\n```\n\n**VS Code** — install the agent plugin (bundles this server):\n\n```bash\ncopilot plugin install desktop-commander/remote-desktop-commander\n```\n\nOr add the server directly to `.vscode/mcp.json` in your workspace (or your user-level `mcp.json`):\n\n```json\n{\n  \"servers\": {\n    \"remote-desktop-commander\": {\n      \"type\": \"http\",\n      \"url\": \"https://mcp.desktopcommander.app/mcp\"\n    }\n  }\n}\n```\n\n**Gemini CLI:**\n\n```bash\ngemini extensions install https://github.com/desktop-commander/remote-desktop-commander\n```\n\n### 3. Try it\n\nThese work against any machine running the agent, no setup needed:\n\n> \"List the contents of my home directory and tell me what's there.\"\n>\n> \"Create a folder called ai-test in my Documents folder. Inside it, write a short Python script that prints the date, and run it.\"\n>\n> \"Start a Python REPL on my machine, import pandas, and tell me which version is installed. Keep the session alive.\"\n>\n> \"Search my home directory for files modified in the last 24 hours and group them by folder.\"\n\nFull walkthrough with troubleshooting: [docs/SETUP.md](docs/SETUP.md)\n\n## Requirements\n\n- Node.js 18 or newer (for `npx`) on each computer you want to control\n- One terminal running the device agent per machine\n- Works on **macOS, Windows, and Linux**\n\n## Available tools\n\n| Category | Tools |\n|---|---|\n| **Read & explore** | Read File, Read Multiple Files, List Directory, Get File Info |\n| **Search** | Start Search, Get More Search Results, Stop Search, List Active Searches |\n| **Write & edit** | Write File, Edit File Block, Move or Rename File, Create Directory, Write or Modify PDF |\n| **Terminal & processes** | Start Process, Send Input to Process, Read Process Output, List Active Sessions, List System Processes, Terminate Session, Kill System Process |\n| **Devices & account** | List Paired Devices, Current User Info, Ping Device, Shutdown Device Agent |\n| **Configuration & help** | Get Configuration, Set Configuration Value, Get Prompt Library, Get Recent Tool Activity, Get Usage Statistics, Submit Feedback |\n\n## How it works\n\n```mermaid\nsequenceDiagram\n    participant AI as AI app (web, mobile, desktop)\n    participant S as mcp.desktopcommander.app\n    participant D as Your computer (device agent)\n    AI->>S: MCP tool call (OAuth 2.0 bearer)\n    S->>D: relayed to your paired device\n    D->>D: executes locally\n    D->>S: result\n    S->>AI: MCP response\n```\n\n- Authentication is standard **OAuth 2.0 with PKCE**; devices pair via the OAuth **device authorization flow** (the short-code approval you see at setup).\n- The server only accesses data explicitly requested through tool calls; transmission is encrypted in transit (HTTPS/TLS).\n- 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](https://mcp.desktopcommander.app).\n- 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](SECURITY.md).\n\n## Support & legal\n\n- **Setup guide:** [docs/SETUP.md](docs/SETUP.md)\n- **Bugs & feature requests:** [GitHub Issues](https://github.com/desktop-commander/remote-desktop-commander/issues) · **Contributing:** [CONTRIBUTING.md](CONTRIBUTING.md)\n- **Community:** [Discord](https://discord.gg/kQ27sNnZr7)\n- **Security reports:** see [SECURITY.md](SECURITY.md)\n- **Privacy policy:** [Remote Desktop Commander privacy policy](https://legal.desktopcommander.app/#/remote_privacy_policy) · **Terms of service:** [legal.desktopcommander.app](https://legal.desktopcommander.app/#/terms_of_service)\n- **License:** [LICENSE](LICENSE) — this repository holds docs and manifests only; the hosted service is proprietary\n\n## Related\n\n- [DesktopCommanderMCP](https://github.com/wonderwhy-er/DesktopCommanderMCP) — the open-source local MCP server (npm: [`@wonderwhy-er/desktop-commander`](https://www.npmjs.com/package/@wonderwhy-er/desktop-commander)), which also ships the `remote` device agent used for pairing\n- [desktopcommander.app](https://desktopcommander.app) — product site\n",
  "bytes": 7534,
  "sha": "5323620f9817e504002ffd4fc0e1e8b4594b9b37762a30b162ceb72c2cbdb290",
  "repo_slug": "desktop-commander/remote-desktop-commander",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_desktop_commander_remote_desktop_command_55948e08/readme"
}