Agent Companion helps you control coding agents running on your own computer from your phone. Start tasks, monitor sessions, approve actions, continue the same thread, and preview local builds, without SSH, Tailscale, Termux, or a cloud-only workflow.

AppSignal — Full-stack monitoring for errors, metrics, and logs

Full-stack monitoring for errors, metrics, and logs

Why I built this I can't always be at my laptop. But I still want to vibe code, kick off tasks, follow up on sessions. Right now, there's no clean way to use your coding agents from your phone. You're stitching together SSH, Tailscale, Termux, and a bunch of other tools that's alot of hassle. I wanted a simple, model-agnostic way to run and control my coding agents from anywhere just my phone. Nothing else. So I built Agent Companion. It's an MVP. It has bugs. It needs iterations. I know all of that. But this is honestly the project I've poured the most thought and effort into from the pairing flow to the session syncing to the approval system and I'd rather get it into people's hands now than keep polishing it alone. There's a long list of things I want to add full CLI experience, more agent support, better UI, tighter security and if this gains traction and people actually find it useful, I'll open source it too. Try it out, and let me know what you think.

@nakshjainsonigara How are you handling approval state once a session hops between laptop and phone, especially with the model agnostic setup? The pairing flow gets you in, but session syncing plus resumable approvals feels like the part that makes Agent Companion safe enough to trust on the go.

AppSignal — Full-stack monitoring for errors, metrics, and logs

Full-stack monitoring for errors, metrics, and logs

Agent Companion

Your coding agents, untethered.

Your coding agents, untethered.