>_ blog

Claude Code Remote Control

Continue local Claude Code sessions from your phone, tablet, or any browser. Your full dev environment stays local while you work from anywhere.

February 2026 · Research Preview · Pro & Max plans

~/projects/rabapl
$ claude remote-control
Starting Remote Control session...
Session URL: https://claude.ai/code/session/abc123
Press space to show QR code
Connected from iPhone via Claude app

>_What is it?

Remote Control connects claude.ai/code or the Claude mobile app to a Claude Code session running on your machine. Start a task at your desk, then pick it up from your phone on the couch or a browser on another computer.

The key insight: nothing moves to the cloud. Your filesystem, MCP servers, tools, and project configuration all stay local. The web and mobile interfaces are just a window into your running session.

💻

Full local environment

Filesystem, MCP servers, project config - all stay available remotely.

🔄

Multi-device sync

Send messages from terminal, browser, and phone interchangeably. Conversation stays in sync.

🔌

Auto-reconnect

Laptop sleeps or network drops? Session reconnects automatically when your machine comes back.

>_How to use it

Option 1: New session

Navigate to your project directory and run:

claude remote-control

The process stays running in your terminal, waiting for remote connections. It displays a session URL and you can press spacebar to show a QR code for quick phone access.

--verbose--sandbox--no-sandbox

Option 2: From an existing session

Already in a Claude Code session? Use the slash command:

/remote-control  # or /rc for short

This carries over your current conversation history and displays the session URL and QR code. Pro tip: use /rename first to give the session a descriptive name so it's easy to find across devices.

Connecting from another device

  • Open the session URL in any browser to go directly to the session on claude.ai/code
  • Scan the QR code to open it directly in the Claude mobile app
  • Browse claude.ai/code and find the session by name (shows a computer icon with green dot when online)

>_How it works

Your local Claude Code session makes outbound HTTPS requests only and never opens inbound ports on your machine. When you start Remote Control, it registers with the Anthropic API and polls for work. When you connect from another device, the server routes messages between the client and your local session over a streaming connection.

All traffic travels through the Anthropic API over TLS - the same transport security as any Claude Code session. The connection uses multiple short-lived credentials, each scoped to a single purpose and expiring independently.

architecture
Remote Control data flow:
Your machine ---> Anthropic API <--- Phone / Browser
(outbound HTTPS) (TLS streaming) (claude.ai/code)
No inbound ports. No code leaves your machine.

>_Remote Control vs Claude Code on the Web

 Remote ControlCode on the Web
Runs onYour machineAnthropic cloud
Local filesFull accessCloned repo only
MCP serversAvailableNot available
SetupTerminal must stay openZero setup
Best forContinuing local work remotelyQuick tasks, parallel sessions

>_Limitations

  • One remote session at a time per Claude Code instance
  • Terminal must stay open - closing the terminal or stopping the process ends the session
  • ~10 minute timeout on extended network outages (machine awake but offline)
  • Pro or Max plan required - not available on Team, Enterprise, or API key auth

>_My take

This is one of the features I didn't know I needed until I tried it. I regularly start long-running Claude Code tasks at my desk and want to check in on progress from my phone. Before Remote Control, I'd either leave the terminal visible and glance at it, or SSH in from my phone - neither great.

Now I just /rc from an active session, scan the QR code with my phone, and I'm in. The conversation syncs both ways, so I can send follow-up instructions from my phone while the terminal keeps showing tool activity. Particularly useful for deploy workflows - kick off a build, walk away, approve the PR from the Claude app when it's ready.

The security model is solid. No inbound ports, no code leaving your machine, short-lived credentials. It's essentially the same trust model as a normal Claude Code session, just with an extra routing layer through the Anthropic API.

Official documentation

Full setup guide, CLI reference, and security details.

Read the docs