What is Claude Code Remote Control?
Claude Code Remote Control lets you continue a Claude Code session that is running on your computer from your phone, a tablet or any browser. You start a task at your desk, then follow and steer it from claude.ai/code or the Claude mobile app.
The key detail, per the Remote Control docs, is that "Claude keeps running locally the entire time." Your filesystem, MCP servers, tools and project configuration stay available, and typing @ on your phone autocompletes file paths from your local project. The web and mobile apps are just a window into that local session.
What you can do from another device:
- Send messages from your terminal, browser and phone interchangeably, with the conversation in sync everywhere.
- Watch subagents and workflows that are already running, and stop them.
- Attach photos and files. Photos go straight to Claude; other files download to your machine as
@file references. - Pick a model or effort level for the session.
- Get push notifications when a long task finishes or Claude needs a decision.
What do you need to use Remote Control?
The requirements are strict, so check these first:
- Plan: Pro, Max, Team or Enterprise. API keys are not supported.
- Team and Enterprise: an Owner must turn on the Remote Control toggle in Claude Code admin settings. It is off by default.
- Sign-in: run
/loginand sign in through claude.ai. IfANTHROPIC_API_KEYorANTHROPIC_AUTH_TOKENis set, it takes precedence, so remove it. - API endpoint: Remote Control only works when Claude Code talks to
api.anthropic.comdirectly. It is not available on Amazon Bedrock, Google Cloud's Agent Platform or Microsoft Foundry, or whenANTHROPIC_BASE_URLpoints at an LLM gateway or proxy. - Feature flags: setting
CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFICorDISABLE_GROWTHBOOKmakes Remote Control unavailable.
A long-lived token from claude setup-token won't work either, because it can only make model requests. If something fails, claude doctor shows which eligibility check failed.
How do you start a Remote Control session?
There are three ways from the CLI.
Server mode keeps a process running in your terminal that waits for remote connections:
claude remote-control
It prints a session URL and, if you press the spacebar, a QR code for your phone. Server mode can host several sessions at once. Useful flags include:
| Flag | What it does |
|---|---|
--name "My Project" |
Sets the session title shown at claude.ai/code |
--spawn worktree |
Gives each new session its own git worktree, so sessions don't edit the same files |
--spawn session |
Serves exactly one session and rejects others |
--capacity <N> |
Maximum concurrent sessions, 32 by default |
--permission-mode <mode> |
Starting permission mode for the server's sessions |
-c, --continue |
Brings back the session the last server in this folder started with |
Interactive mode gives you a normal terminal session that is also available remotely:
claude --remote-control "My Project"
From a session you already have open, run the command and your conversation history comes with it:
/remote-control
/rc is the short alias. The first time, Claude Code asks you to confirm before it connects. The same /remote-control command works in the VS Code extension and in local sessions in the Claude Desktop app. The flags are listed in the CLI reference.
How do you connect from your phone?
Once a session is live, you have three options:
- Open the session URL in any browser.
- Scan the QR code to open the session in the Claude app.
- Open claude.ai/code or the Claude app and find the session by name. In the mobile app, tap Code. Online Remote Control sessions show a computer icon with a green dot.
No app yet? Run /mobile inside Claude Code for a QR code that opens the right app store.
To check the connection, look for the /rc active indicator in your terminal, or run /remote-control again to open a status panel with the URL, the QR code and a disconnect option.
Turn on push notifications
Install the Claude app, sign in with the same account you use in the terminal, allow notifications, then run /config in Claude Code and enable Push when Claude decides, Push when actions required, or both. Claude Code skips pushes while you're typing in the connected terminal.
Connect every session automatically
By default, Remote Control only starts when you ask for it. To connect every interactive session, run /config and set Enable Remote Control for all sessions to true, or set remoteControlAtStartup to true in ~/.claude/settings.json. A checked-in project settings file can turn auto-connect off for a repo but can't turn it on.
Is Claude Code Remote Control secure?
Per the docs, your local session "makes outbound HTTPS requests only and never opens inbound ports on your machine." It registers with the Anthropic API and polls for work, and all traffic travels over TLS using short-lived credentials, each scoped to one purpose.
Two things to know:
- Transcripts are stored. While Remote Control is connected, your messages, Claude's responses and tool activity are stored on Anthropic servers to keep devices in sync. Execution and file access stay on your machine.
- It can be turned off. The
disableRemoteControlsetting blocks it entirely, and organizations with Zero Data Retention or the HIPAA configuration can't enable it.
For extra protection, Trusted Devices (in beta) requires each browser, phone or desktop app to be enrolled and the sign-in to be no more than 18 hours old. After that, members confirm with Face ID, Touch ID, Windows Hello or a passkey. Owners turn it on for Team and Enterprise organizations; Pro and Max users can turn on Require trusted devices in their own settings.
Remember that a remote message is still you giving Claude instructions on your machine. Your permission settings apply as usual, and permission prompts are forwarded to your device for you to answer.
What are the limits of Remote Control?
- The process must keep running. Closing the terminal, quitting the app or stopping
claudetakes the session offline. On a remote server over SSH, start Claude Code insidetmuxorscreen. - One remote session per interactive process. Use server mode for several at once.
- Network outages. In server mode, Claude Code gives up after roughly 10 minutes without network and exits. An interactive session keeps retrying until the network returns.
- Some commands are local only.
/pluginand/resumework only in the terminal. Commands like/compact,/clear,/contextand/usagework from mobile and web, and/model sonnetor/effort highwork when you pass the value as an argument. - Resuming. After you stop server mode with Ctrl+C, running
claude remote-controlagain in the same folder brings its sessions back for about four hours.
Remote Control vs cloud sessions: which should you use?
Both use the claude.ai/code interface. The difference is where Claude runs:
| Remote Control | Cloud session | |
|---|---|---|
| Runs on | Your computer | Anthropic-managed cloud by default |
| Your local files, MCP servers and tools | Yes | No, it starts from a fresh clone |
| Computer must stay on | Yes | No |
| Best for | Continuing local work from another device | Starting tasks without local setup, or running several in parallel |
The Claude Code on the web guide covers cloud sessions. If you want work to start on its own at a set time rather than steering it yourself, see Claude Code scheduled tasks.
Bring your prompts with you
Typing long instructions on a phone keyboard is slow. Kipt keeps your prompts as Markdown in a private GitHub repo you own, with {{name}} blanks for the parts that change, so you can fill in a saved prompt and paste it into a Remote Control session instead of retyping it.