Initialize squad on status-call server side respon...
# support
p
Hello, I think i'm missing something as, while my server replies, the squad doesn't seem to be initialized. on the vapi side, i have no squad created and an empty assistant to initialize the call whose phone number is linked to. currently my server replies with: {"squad":{"members":[{"assistant_id":"5f4faadd-5888-4ad7-a504-8f3035830b94","assistant_destinations":[{"type":"assistant","assistant_name":"faq","message":" ","description":"Transfer à l'assistant de faq"},{"type":"assistant","assistant_name":"spelling","message":" ","description":"Transfer à l'assistant de spelling"},{"type":"assistant","assistant_name":"booking","message":" ","description":"Transfer à l'assistant de booking"}]},{"assistant":{"name":"faq","assistant_id":"0c2fd21f-d0bf-4353-a76d-0348b879cb75"}},{"assistant":{"name":"spelling","assistant_id":"beecc680-7d5b-4c2c-bc1f-771dbc0aefdf"}},{"assistant":{"name":"booking","assistant_id":"67ed59f3-2ad6-44b2-b244-4e4f0a4ac83f"}}]}} But it doesn't trigger the squad as no body is speaking on the other side but an default assistant . Any input appreciated
Ok apaprently i need to reply to assistant request not status update.
I don't see assistant-request call type in the list of calls
neither i do on server messages options in the advanced assistant tab
as per the documentation: https://docs.vapi.ai/squads-example it should be an assistant-request call to the server, am I correct?
maybe i should ha no assistant connected to the phone number so it does the assistant-request?
i noticed: Vapi will send 'assistant-request' to your Server URL (https://expertsassistantsapi.fly.dev/api/v1/vapi) to fetch assistant.
how can i test it in the web browser if it only works when i call the number?
s
Yes your correct and it's not available in the assistant section instead you have to subcribe to it by setting the server url to the phone number.
Yes it only works for the phone number calls.
@paul gaston do let me know if you have more doubts.
p
i ended up making the whole thing working
the core idea was the destribtion of the transition and appropriate prompt
now I have another issue, the assistant work flow seems to work but at some point the 2nd one hangs up the phone without reason and the end of call server message doesn"t have any errors
only coast breakdown: "ended_reason":"pipeline-error-anthropic-llm-failed","cost":0.1589,"cost_breakdown":{"stt":0.0086,"llm":0.0994,"tts":0.0056,"vapi":0.0406,"total":0.1589,"llm_prompt_tokens":31881,"llm_completion_tokens":253,"tts_characters":375,"analysis_cost_breakdown":
are you familiar with this issue?
s
can you share the call id? and tag me to with next your next messages so i can reicieve a notification and reply to you faster.
5 Views