> For the complete documentation index, see [llms.txt](https://docs.remote.it/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.remote.it/mcp-server/mcp.md). # MCP Server Overview The Remote.It MCP server lets AI assistants like Claude work with your Remote.It account directly — inventory your devices, open and close connections, manage services and networks, and (when enabled) run scripts on your fleet, all through natural conversation. It is a hosted [Model Context Protocol](https://modelcontextprotocol.io) (MCP) server: there is nothing to install or run locally. You add it to your AI assistant as a connector, sign in with your Remote.It account, and the assistant gains a set of Remote.It tools it can use on your behalf. {% hint style="warning" %} **Preview release.** The MCP server is currently in preview at: ``` https://mcp.demo.remote.it/mcp ``` This URL will change when the server is fully released. Check back here for the final address. {% endhint %} ## What can it do? Once connected, you can ask your AI assistant things like: * "Which of my devices are offline right now?" * "Connect to SSH on my office Raspberry Pi" * "How much bandwidth did my gateway use this week?" * "Tag all my lab devices and share them with " * "Run the update script on every device tagged `production`" The assistant translates these into calls to Remote.It tools — the same operations available in the [web portal](https://app.remote.it) and [Desktop app](/software/desktop.md), scoped to your account and permissions. See the [Tool Reference](/mcp-server/tools.md) for the full list. ## How access works * **Sign in with OAuth — no API keys.** When your assistant first uses the connector, you sign in through your browser with your normal Remote.It account (including social sign-in and 2FA). You never paste keys or secrets into the assistant. * **You choose what to grant.** The consent screen lets you limit what the assistant can do — for example, read-only access to devices without the ability to connect or make changes. See [Permissions & Safety](/mcp-server/permissions-and-safety.md). * **Your existing permissions still apply.** The assistant can never do more than your Remote.It account can. Organization roles and device permissions are enforced on every call. * **Destructive actions are confirm-gated.** Deleting devices, running scripts, and similar operations require an explicit confirmation step. ## Get started You'll need: * **A Remote.It account** — [sign up for a free account](https://app.remote.it/#/sign-up) that will let you access up to 5 devices. * **An MCP-capable AI assistant**, such as Claude. Then: * [Connect Claude](/mcp-server/connect-claude.md) — add the connector to Claude on web, desktop, mobile, or Claude Code * [Tool Reference](/mcp-server/tools.md) — every tool the server exposes * [Permissions & Safety](/mcp-server/permissions-and-safety.md) — scopes, consent, and safety gates The server works with any MCP client that supports the Streamable HTTP transport and OAuth 2.1 authorization, so other MCP-capable apps can connect the same way. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.remote.it/mcp-server/mcp.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.