12:44:11 [agent/embedded] embedded run agent end: runId=0eebf6a6-24a6-4590-9b03-97d0b75569a9 isError=true model=gpt-5-mini provider=github-copilot error=LLM request failed: provider rejected the request schema or tool payload. rawError=400 {"message":"","code":"invalid_request_body"}
12:44:11 [agent/embedded] auth profile failure state updated: runId=0eebf6a6-24a6-4590-9b03-97d0b75569a9 profile=sha256:bbb6682dd900 provider=github-copilot reason=format window=cooldown reused=false
12:44:11 [agent/embedded] embedded run failover decision: runId=0eebf6a6-24a6-4590-9b03-97d0b75569a9 stage=assistant decision=surface_error reason=format from=github-copilot/gpt-5-mini profile=sha256:bbb6682dd900
#Discord Bot stops responding
1 messages ยท Page 1 of 1 (latest)
Let me check the docs and source for that exact failure path.
That looks like a model request failure, not the Discord bot transport dying.
github-copilot/gpt-5-mini returned 400 invalid_request_body, and OpenClaw classified it as a schema/tool-payload format error. The surface_error bit usually means there was no usable fallback after that failure.
Can you paste these first:
openclaw --version
openclaw status --all
openclaw models status --probe
Then reproduce once with:
openclaw logs --follow