Claude Code Remote Control: How to Use It From Your Phone (2026)

Last Updated: November 6, 2026

Summary: Claude Code Remote Control lets you keep driving a Claude Code session that runs on your own computer from your phone, tablet or any browser. Start it with claude remote-control, claude --remote-control or /remote-control, then open the session at claude.ai/code or in the Claude app. It works on Pro, Max, Team and Enterprise plans, costs nothing extra, and your code never leaves your machine.

You kick off a long refactor at your desk, then have to leave for a meeting. Before February 2026 that meant either waiting or killing the session. Anthropic's answer is Remote Control, a Claude Code feature that turns your phone into a window onto the session still running on your laptop.

This guide covers what Remote Control is, which plans include it, every way to start it, how to connect from your phone, how the security model works, and the limits worth knowing before you rely on it. Commands and requirements were checked against Anthropic's official Claude Code documentation in November 2026.

Claude Code Remote Control connects claude.ai/code or the Claude mobile app (iOS and Android) to a Claude Code session running on your own computer. Claude keeps running locally the whole time. Your phone or browser only sends messages and shows what is happening, so code execution, file access and your tools stay on your machine.

Anthropic announced the feature on February 24, 2026, as a research preview for Max subscribers, according to Gigazine, with Pro access promised soon after. The official documentation now lists Pro, Max, Team and Enterprise as supported plans.

With a session connected, you can:

New to the tool itself? Start with our explainer on what Claude Code is, then come back here.

Remote Control is included at no extra cost on every paid Claude subscription that includes Claude Code. It does not work with API keys, and it is off by default for organizations until an Owner turns it on.

If you route Claude Code through an LLM gateway or proxy by setting ANTHROPIC_BASE_URL to another host, Remote Control also stops working, even if you are signed in with claude.ai. For more on what each plan's limits look like in practice, see our guide to Claude API pricing.

There are three ways to start Remote Control from the terminal, plus a slash command in the Claude Desktop app and the VS Code extension. Pick based on whether you are starting fresh or already mid-task.

In your project folder, run claude remote-control. The first time, it explains what it does and asks you to confirm. The process then waits in your terminal for remote connections, shows a session URL, and displays a QR code when you press the spacebar.

Server mode is the most flexible option. Useful flags include --name "My Project" to set the title you see on your phone, --spawn worktree to give each new session its own git worktree so parallel sessions don't edit the same files, --capacity to cap concurrent sessions (32 by default), and --permission-mode to set a starting permission mode such as acceptEdits.

Run claude --remote-control, optionally followed by a name in quotes. You get a normal Claude Code session in your terminal that is also reachable from claude.ai and the Claude app. Unlike server mode, you can keep typing locally.

Type /remote-control or /rc inside a running session. Remote Control picks up your current conversation history, so you can walk away mid-task without starting over. While it is connected, the terminal shows an /rc active indicator. Run the command again to see the URL and QR code or to disconnect.

Want it on all the time? Run /config and set Enable Remote Control for all sessions to true. In the Desktop app the same switch is under Settings, then Claude Code.

Once a session is live, you have three ways in:

From your phone you can send messages, approve permission prompts, answer Claude's questions, review a diff of your changes when the project is a git repository, switch models with /model, change effort with /effort, and run commands such as /compact, /clear, /context and /usage. Some terminal-only commands, including /plugin and /resume, still need the local CLI.

To turn on push notifications, run /config in your terminal and enable Push when Claude decides, Push when actions required, or both. You can also ask for one in your prompt, for example "notify me when the tests finish."

Remote Control is designed so your machine never accepts incoming connections. Your local session makes outbound HTTPS requests only and opens no inbound ports. It registers with the Anthropic API, and messages between your phone and your computer are routed through Anthropic over TLS using several short-lived credentials, each limited to one purpose.

The trade-off is the transcript. While Remote Control is connected, your messages, Claude's replies and tool activity are stored on Anthropic's servers so the conversation can sync across devices and recover from network drops. Execution and file access still stay local. That is why organizations with Zero Data Retention can't enable Remote Control, and HIPAA-configured organizations see the toggle greyed out.

For extra protection, Anthropic offers Trusted Devices (in beta). When it is on, viewing or steering a Remote Control session requires an enrolled device and a sign-in no older than 18 hours, which you can refresh with Face ID, Touch ID, Windows Hello or a passkey. Anthropic says it never receives biometric data, only each device's public key and basic metadata. Pro and Max users switch it on in their own settings, and Team and Enterprise Owners turn it on for the whole organization. The full setup steps are in Anthropic's Trusted Devices documentation.

IT teams can also block the feature on specific machines with the disableRemoteControl managed setting. A checked-in project settings file can turn auto-connect off for a repository but can never turn it on for everyone who opens it.

Both use the same claude.ai/code interface, which causes confusion. The difference is where the work happens. Anthropic's guide to Claude Code on the web covers the cloud option in detail.

Anthropic also offers Dispatch, which lets you message a task from the Claude app and have the Desktop app spin up a session for it, and Channels, which push events from apps such as Telegram or Discord into a session. Remote Control is the one built for steering work that is already in progress.

For a solo developer, Remote Control is a convenience. For a business paying for Claude seats, it changes how much value each seat produces, because long agent runs no longer stall the moment someone steps away from their desk.

Claude Code is already central to how many teams build software. Our Claude Code statistics page tracks its revenue and adoption, and the Claude Code vs Cursor comparison shows how it stacks up against the leading AI editor.

For anything else, claude doctor shows which eligibility check failed, and claude remote-control --verbose prints the full connection log.

What is Claude Code Remote Control?
Remote Control is a Claude Code feature that lets you continue a session running on your own computer from your phone, tablet or any browser, through claude.ai/code or the Claude mobile app. Claude keeps running locally, so your files, tools and MCP servers stay on your machine while you send messages and approve actions remotely.

Is Claude Code Remote Control free?
There is no extra charge for Remote Control, but it requires a paid Claude plan: Pro at $20 a month, Max, Team or Enterprise. It does not work on the free plan or with API keys. On Team and Enterprise, an Owner must enable it in the Claude Code admin settings first.

How do I use Claude Code from my phone?
Run claude remote-control in your project folder, or type /remote-control inside an existing session. Then scan the QR code with your phone, open the session URL, or find the session in the Code tab of the Claude app. Your computer must stay on while you work.

Does my computer need to stay on for Remote Control?
Yes. Remote Control sessions run on your own machine, so the computer has to be on and the claude process has to keep running. If your laptop sleeps or loses Wi-Fi briefly, the session reconnects when it comes back. To work without your computer, use Claude Code on the web, which runs in the cloud.

Is Claude Code Remote Control secure?
Your machine only makes outbound HTTPS connections and never opens inbound ports, and all traffic goes through the Anthropic API over TLS with short-lived credentials. The session transcript is stored on Anthropic's servers while connected, so Zero Data Retention organizations can't use it. Trusted Devices adds device enrollment and biometric re-verification.

What is the difference between Remote Control and Claude Code on the web?
Remote Control runs Claude on your own computer, with your local tools and files, and your phone acts as a remote screen. Claude Code on the web runs sessions in the cloud, so you need no local setup and your computer can be off, but your local MCP servers and configuration are not available.

Claude Code Remote Control solves a simple but costly problem: AI coding agents that stall the moment you leave your desk. One command, claude remote-control or /rc, puts a running local session in your pocket without moving your code to the cloud.

If you are on Pro or Max, you can try it today. If you run a Team or Enterprise account, turn it on alongside Trusted Devices and give your developers a way to keep long tasks moving from anywhere.

Written by the AI Business Weekly Editorial Team. Reviewed, edited and published by Sameer Khan. This article was AI-assisted, then reviewed by Sameer Khan before publishing.