Claude Code Remote Control: Run Sessions From Your Phone
Claude Code Remote Control lets you drive a session on your own computer from your phone or a browser. Plans, commands, server mode, security and limits.
Josh Pocock8 min read
AI Coding
Remote Control
Stride AI Academy
Claude Code Remote Control connects the Claude mobile app or claude.ai/code to a Claude Code session running on your own computer. You start a task at your desk, then keep steering it from your phone or another browser. Claude keeps running locally the whole time, so your files, tools and MCP servers stay on your machine. Start it with claude remote-control in your project, or /remote-control inside a session you already have open.
This guide covers who can use it, the three ways to start it, how to connect your phone, what you can and can't do from mobile, security, and how it differs from Claude Code on the web. Everything here was checked against Claude Code's Remote Control docs on November 22, 2026.
What Remote Control is
Remote Control makes your phone or browser a window into a local session. The docs list what that gets you:
- Your full local environment. Your filesystem, MCP servers, tools and project config all stay available, and typing
@autocompletes file paths from your local project. - Both surfaces at once. The conversation, subagents and workflows stay in sync, so you can type from the terminal, browser and phone interchangeably.
- Photos and files from your phone. Claude sees attached photos directly. Other files are downloaded to your machine and passed to Claude as
@file references. - Reconnects. If your laptop sleeps or the network drops, Claude Code reconnects when the machine comes back online.
The catch is in the same sentence as the benefit: your computer has to stay on and the claude process has to keep running.
Which plans have Claude Code Remote Control
On November 22, 2026, the docs list these requirements:
| Requirement | Detail |
|---|---|
| Plan | Pro, Max, Ultra, 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 first. |
| Sign-in | Sign in through claude.ai with /login. |
| Provider | Not available on Amazon Bedrock, Google Cloud's Agent Platform or Microsoft Foundry, or when ANTHROPIC_BASE_URL points at a gateway or proxy. |
| Feature flags | Unavailable if you set CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC or CLAUDE_CODE_DISABLE_FEATURE_FLAGS. |
Organizations with Zero Data Retention or the HIPAA configuration can't enable it. If you're choosing a plan, our Claude pricing guide compares them.
How to start a Remote Control session
There are three ways from the CLI. The desktop app and the VS Code extension use the slash command.
Server mode
Run this in your project directory:
claude remote-controlclaude rc is the short alias. The first time, it asks Enable Remote Control? (y/n). The process then waits for remote connections, shows a URL, and shows a QR code when you press the spacebar.
Useful flags from the docs:
| 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 |
--spawn session |
Serves exactly one session and rejects others |
--capacity <N> |
Maximum concurrent sessions, default 32 |
--permission-mode <mode> |
Starting permission mode, such as acceptEdits |
-c, --continue |
Brings back the session the last server in this folder started |
The default spawn mode, same-dir, puts every session in the same folder, so they can clash if two edit the same files. Press w while the server runs to switch between same-dir and worktree. Our worktrees guide explains why separate checkouts help.
An interactive session with Remote Control on
claude --remote-control "My Project"--rc works too. You get a normal terminal session that you can also drive from your phone, and the name is optional.
From a session you already have open
/remote-control/rc is the alias. This carries your current conversation over, so it's the one to use when you're mid-task and about to leave your desk. Run it again to see the URL and QR code, or to disconnect while the local session keeps running.
How to connect from your phone
Once a session is live, the docs give three ways in:
- Open the URL Claude Code shows in any browser.
- Scan the QR code to open the session in the Claude app.
- Open claude.ai/code or the Claude app and pick the session from the list. In the mobile app, tap Code. Remote Control sessions show a computer icon with a green dot when online.
If you don't have the app yet, run /mobile inside Claude Code. It shows a QR code for claude.ai/mobile, which opens the right app store.
Turn it on for every session
Run /config and set Enable Remote Control for all sessions to on. In a settings file, the key is remoteControlAtStartup:
{
"remoteControlAtStartup": true
}The docs say a true only counts in your user ~/.claude/settings.json or managed settings. A checked-in project file can turn it off for a repo but can't turn it on for everyone.
Get push notifications
In /config, turn on Push when Claude decides, Push when actions required, or both. Sign in to the mobile app with the same account and allow notifications. You can also ask for one in a prompt, such as notify me when the tests finish.
What works from mobile and the web
Most of the session works remotely: messages, answers to permission prompts, model and effort changes, a diff pane for git repos, and stopping subagents. Some commands only run in the local terminal. The docs name /plugin and /resume as local-only. These work from mobile and web:
/compact,/clear,/context,/usage,/exit,/usage-credits,/recapand/reload-plugins/model,/effort,/fast,/colorand/rename, with the value as an argument, such as/model sonnet
You can also fork a session at claude.ai/code to try another direction while the original keeps running.
Is Remote Control secure?
The docs describe the connection this way:
- Your machine makes outbound HTTPS requests only and never opens inbound ports.
- All traffic goes through the Anthropic API over TLS, using several short-lived credentials, each scoped to one purpose.
- While connected, the transcript is stored on Anthropic's servers so devices stay in sync. Code runs and files stay on your machine.
For extra protection, Pro, Max and Ultra users can turn on Require trusted devices in their settings. Team and Enterprise Owners turn it on for the organization. It requires an enrolled device and a sign-in no more than 18 hours old, confirmed with Face ID, Touch ID, Windows Hello or a passkey. To turn Remote Control off completely, use the disableRemoteControl setting.
Permission prompts still appear, and the docs say they stay open until you answer them, from the terminal or your phone. If you want fewer prompts, see our guides to Claude Code permissions and Claude Code sandboxing.
Remote Control vs Claude Code on the web
Both use the claude.ai/code interface. The difference is where the session runs.
| Remote Control | Claude Code on the web | |
|---|---|---|
| Runs on | Your computer | Cloud machines, Anthropic-managed by default |
| Local MCP servers and tools | Available | Not available |
| Computer must stay on | Yes | No |
| Best for | Steering work you started locally | Tasks with no local setup, repos you haven't cloned, parallel jobs |
Use Remote Control when you're mid-task and want to keep going from the couch. Use a cloud session when you want to start something without your laptop. Our Claude Code on the web guide covers that side. For work that should start on a timer rather than when you pick up your phone, use Claude routines.
Limits to know
From the docs' limitations list:
- One remote session per interactive process. Use server mode for several at once.
- The local process must keep running. Close the terminal and the session goes offline. On a remote machine over SSH, start
claudeinsidetmuxorscreen. - Network outages. In server mode Claude Code gives up after about 10 minutes and exits. An interactive session keeps retrying until the network returns.
- Resume window. After you stop a server with Ctrl+C,
claude remote-controlin the same folder brings its sessions back for about four hours. - Other dialogs expire. Permission prompts and questions wait for you. Other forwarded dialogs close after five minutes by default, set with
dialogExpiry.
Learn to build with Claude Code
The Stride AI Coding course teaches Claude Code and Codex for building client tools, written for non-programmers, with weekly live calls to review your builds. Many of the free guides cover Claude Code setups, and the Claude Code community is where members compare theirs.
FAQ
What is Claude Code Remote Control?
It connects the Claude mobile app or claude.ai/code to a Claude Code session running on your own computer. You can send messages, answer permission prompts and watch progress from your phone while the code and files stay on your machine.
Is Claude Code Remote Control free?
It isn't on the Free plan. The docs list Pro, Max, Ultra, Team and Enterprise, and API keys aren't supported. On Team and Enterprise, an Owner has to switch it on first.
How do I use Claude Code on my phone?
Run claude remote-control or /remote-control on your computer, then scan the QR code or open the session from the Code tab in the Claude app. Run /mobile if you need the app first.
Does my computer need to stay on?
Yes. Remote Control sessions run on your machine, so the computer has to be on and the claude process running. If you want work to run with your laptop closed, use Claude Code on the web or routines instead.
What's the difference between Remote Control and Claude Code on the web?
Remote Control runs on your computer with your local tools and MCP servers. Claude Code on the web runs in the cloud, so it needs no local setup but can't reach your local environment.
Can I run several Remote Control sessions at once?
Yes, in server mode. claude remote-control serves up to 32 sessions by default, set with --capacity. Add --spawn worktree so each session gets its own git worktree.