My agent is not responding. | Friends of the Crustacean 🦞🤝 | Page 1

raw steppe Apr 18, 2026, 10:22 AM

#

10:18:13 [agent/embedded] incomplete turn detected: runId=9eac0373-459e-4993-96ac-ccccdfe48a19 sessionId=767f838c-8833-41bb-9deb-be9a00a6dc13 stopReason=stop payloads=0 — surfacing error to user

tidal helmBOT Apr 18, 2026, 10:22 AM

#

sonic capeBOT Apr 18, 2026, 10:22 AM

#

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

forest fiberBOT Apr 18, 2026, 10:24 AM

#

Let me check the docs and the code path that emits that exact error.

#

I found the exact log line in the runner. I’m also checking the main FAQ path, because the docs layout moved a bit.

#

That line usually means the model run finished, but returned no user-visible reply, so OpenClaw surfaced it instead of silently hanging.

Can you paste these in one message:

openclaw --version
openclaw status --all
openclaw models status

Then reproduce it once while tailing logs:

openclaw logs --follow

Also tell me:

#

which model/provider you’re using
whether it’s cloud or local (Ollama, LM Studio, vLLM, etc.)

Most common causes here are:

older build, there was a recent fix around empty/reasoning-only turns
local OpenAI-compatible backend returning an empty response
backend not handling tools/structured content correctly

Docs:
https://docs.openclaw.ai/help/faq
https://docs.openclaw.ai/gateway/troubleshooting

raw steppe Apr 18, 2026, 10:33 AM

#

10:32:05+00:00 warn gateway/ws {"subsystem":"gateway/ws"} {"cause":"pairing-required","handshake":"failed","durationMs":9,"lastFrameType":"req","lastFrameMethod":"connect","lastFrameId":"33ebd6de-efc4-4045-97ad-024108417944","host":"127.0.0.1:18789","remoteAddr":"127.0.0.1","remotePort":44660,"localAddr":"127.0.0.1","localPort":18789,"endpoint":"127.0.0.1:44660->127.0.0.1:18789","deviceId":"0029c093c85198cb69ff7b0ecad73c04c184a8773a9c407ccc1a759d824ba454","requestId":"a2df7634-7829-4636-8e83-f57abcc7f858","reason":"scope-upgrade"} closed before connect conn=4c4fd0b0-8fd0-46ae-a922-26c5a65d6580 peer=127.0.0.1:44660->127.0.0.1:18789 remote=127.0.0.1 fwd=n/a origin=n/a host=127.0.0.1:18789 ua=n/a code=1008 reason=connect failed
10:32:06+00:00 warn gateway {"subsystem":"gateway"} security audit: device access upgrade requested reason=scope-upgrade device=0029c093c85198cb69ff7b0ecad73c04c184a8773a9c407ccc1a759d824ba454 ip=unknown-ip auth=token roleFrom=operator roleTo=operator scopesFrom=operator.read scopesTo=operator.admin,operator.approvals,operator.pairing,operator.read,operator.talk.secrets,operator.write client=cli conn=e89a13f5-8a30-4972-a732-3af7212f8b0d
10:32:06+00:00 error gateway connect failed: GatewayClientRequestError: pairing required
10:32:06+00:00 error gateway connect failed: GatewayClientRequestError: pairing required
gateway connect failed: GatewayClientRequestError: pairing required

forest fiberBOT Apr 18, 2026, 10:34 AM

#

I found the likely root cause, let me verify the exact recovery steps in the docs so I don’t give you a made-up command.

#

Yep, this is the real blocker.

Your agent issue is not the model first here, it’s Gateway device pairing / scope upgrade.

What the log says:

current device/session has only operator.read
it is requesting higher scopes: operator.admin, operator.approvals, operator.pairing, operator.write
Gateway rejects that with PAIRING_REQUIRED
so the client never fully connects, which is why the agent ends up “not responding”

Run this on the gateway host:

openclaw devices list
openclaw devices approve a2df7634-7829-4636-8e83-f57abcc7f858

#My agent is not responding.