Continue a local Claude Code session from your phone, tablet, or any browser using Remote Control. Works with claude.ai/code and the Claude mobile app.
Free Options
Launch Team
Render Sandboxes: a computer for every agentGive every agent its own machine. Starts in under a second
Promoted
Anthropic announced a new Claude Code feature called Remote Control.
The idea is pretty straightforward: you start a Claude Code session locally in your terminal, then continue them from your phone. Take a walk, see the sun, walk your dog without losing your flow. It's rolling out now to Max users as a research preview.
Anyone here on Max plan tried it yet? How the mobile experience feels in practice (latency, editing capabilities, etc.)?
Reminds me of @Macky which @sayuj_suresh launched a few days ago. Inviting your thoughts on this launch Sayuj. :)
@rohanrecommends thanks for bringing up Macky! yesss this is a good product, but it allows access from mobile for already started sessions but Macky allows you to access your full terminal, so you can do all kinds of terminal stuff and not just Claude.
@sayuj_suresh Ah yes, that matches what I thought the key differentiation was. I was thinking of your product when I saw this and wanted to get your thoughts. Thanks for stopping by and sharing! :)
@rohanrecommends , I'm using openClaw and it sets my expectations pretty high, Now I'm thinking of giving "Claude Code Remote Control" access to my openClaw (OC) connection - so if they need anything there another official from anthropic workflow running in same machine (a extra set of hands).
Apart from it, I'm excited to test it out personally, not sure weather it can control the browser or not but apart from this it looks interesting as I don't need to setup macmini for it.
I don't like this 10 minute limitation as of now but it can be overcome by OC :
"Extended network outage: if your machine is awake but unable to reach the network for more than roughly 10 minutes, the session times out and the process exits. Run claude remote-control again to start a new session."
Just a suggestion : It should keep running until user stops it, ( I mean it won't cost much to keep the connection alive or give a alert like "Your session is ending soon, do you still want to continue "? and instead of starting a new session it should resume the prev. session.
"Take a walk, see the sun, walk your dog without losing your flow" is doing a lot of work in that description and it genuinely lands. The mental cost of losing context mid-session is way higher than people realize until they experience picking up exactly where they left off.
The one remote session at a time limitation is an interesting design choice. Keeps things simple and avoids conflict, but I'm curious if that becomes a friction point when you have multiple long-running jobs going and want to check in on a specific one from your phone.
I use Claude Code daily for building my AI platform and the number of times I've walked away from my desk mid-session wishing I could nudge it or check progress from my phone is higher than I'd like to admit. This is exactly the feature I didn't know was missing.
This is exactly what I always had in my bucket of TODOs, while I travel in bus to University and the way back is in total 3 hours and I believed having something like this would seriously make that time of mine productive.
Been using the @Cursor Agents, @Jules and recently the @OpenClaw to work in that time of the day. This will be the next try but I always thought how about having a computer on the go with me in the bus (not literally but to be able to control one remotely).
Visualising this I had one image like being able to project through maybe some VR glasses and do it? But to be cheap I also thought about how using a mobile maybe we can remotely do this and we're here.
What you'll think about the future ? Let me know and I'll share something I truly believe is the future but so much time for it to happen it seems.
Congrats on the launch! Being able to monitor local sessions from a phone is a huge win for mobility. As we develop ParkEase (a SaaS for property managers), we’re constantly looking for ways to make complex backend tasks manageable from a mobile device.
Quick question: Does the remote control interface support multi-tenant sessions, or is it strictly one agent per remote link? This looks super useful for our dev workflow. Good luck today!
Nice for monitoring long-running tasks. Would be good to get push notifications when it needs approval or hits an error. Or you already support that?
The 10-minute network timeout is the detail I'd want to stress-test first. If you're mid-approval on something destructive and the session drops quietly, that's a bad day. Does it halt the agent on timeout or let it keep running locally without the remote tether?