Connectivity with OpenClaw

issues & bugs
1

I have been trying to use the R1 regularly in combination with my local instance of OpenClaw. The challenge is that nearly half the times, the R1 gets stuck on a spinner after I submit my PTT request.

When diagnosing, the issue seems to be always the same:

  1. rabbit-agent is running
  2. OpenClaw CLI is ok
  3. The gateway is healthy
  4. There is an open ACP session but it’s idle.

The R1 connected, did the initial handshake, but no actual prompts are coming through from Rabbit’s cloud.

Am I doing anything wrong?

EDIT: I created a small skill for OpenClaw to diagnose issues with R1 connectivity btw ( GitHub - GrewingM/r1-triage · GitHub )

2

have a similar issue except with me its hermes that spins even when openclaw works. the fix is generally pressing the action button 5 times or doing a lock wake cycle. After extensive bug tracing, Agent on r1 intermittently enters a stale session state. rabbit-agent successfully launches and initializes Hermes ACP (Initialize → New session), but subsequent r1 input produces no session/prompt and no corresponding write to the stdin FIFO. even when other providers remain operational. After an r1 lock/wake or cloud-session reconnect, rabbit-agent immediately writes to the existing FIFO and processes the prompt normally within seconds. I dont know what rabbit needs to do to fix this, but there not really anything you can do.

3

Thanks, this matches what I traced on my side, and the lock/wake tip is a lighter recovery than the reboot I had been doing.

Same signature here with OpenClaw rather than Hermes: the ACP job starts, initialise and session/new both return, the session list comes back, and then r1 input produces no session/prompt at all. Nothing reaches the job’s stdin.

One deterministic trigger to add: restarting the OpenClaw gateway orphans the session the cloud is holding. After that, the r1 gets at most one reply, and then silence, and no amount of PTT brings it back until the device reconnects.

Two other causes produce the identical symptom but are fixable on the server, worth ruling out before blaming the cloud:

  • rabbit-agent not running at all. The installer starts it with nohup and relies on an @reboot crontab line that is not always written. Mine had been dead for three weeks after a reboot. Check with pgrep -f "rabbit-agent run".
  • OpenClaw 2026.8.x with more than one agent configured. The ACP bridge creates sessions under an unscoped key and the gateway refuses prompts on it, so the r1 sits on “listening” while gateway health says OK.

Wrote both up with fixes here, and the triage script is at GitHub - GrewingM/r1-triage · GitHub if useful.