https://vapi.ai/ logo
Join Discord
Powered by
Enable transient assistant
# support
d
1. on the start of the call, what needs to be enabled to trigger a request to get a transient assistant? I have a make webhook setup that injects info and sends it back. I can't see any incoming requests to the endpoint on the phone call. the server url is set 2. to confirm, there's only 1 server url per assistant?
v
message has been deleted
@DM007 1. To trigger a request to get a transient assistant at the start of the call, you need to ensure that the 
assistant
parameter is set to 
null
and that you are using the 
assistantId
to specify which assistant you want to use for the call. If you have a webhook set up and are not seeing any incoming requests, please double-check that the 
serverUrl
is correctly configured in your VAPI settings. Additionally, ensure that the 
phoneNumber.serverUrlSecret
is properly set if you are using a secret for security. 2. Yes, that's correct. There is only one 
serverUrl
per assistant. This URL is where VAPI will send requests, and it should be configured to handle incoming requests from VAPI and respond appropriately. If you need to use different URLs for different purposes, you would need to set up different assistants, each with their own 
serverUrl
.
You can continue this conversation by mentioning me in the message.
d
@Vapi 1. there is no parameter like that
v
@DM007 I apologize for any confusion. To trigger a request to get a transient assistant at the start of the call, you should ensure that your webhook is correctly configured in your VAPI application settings. The webhook should be set to trigger on the 
answer_url
event, which is called when a call is answered. If you're not receiving requests, please verify the following: - The 
answer_url
is correctly set to your server URL in the VAPI application settings. - Your server is correctly configured to receive POST requests from the VAPI platform. - There are no firewalls or network issues that might be blocking incoming requests to your server. - The server is up and running when the call is initiated. If everything seems to be configured correctly and you're still not seeing incoming requests, it might be best to check the server logs for any errors or contact support for further assistance. Regarding the second point, yes, typically there is only one server URL per assistant that handles the webhook requests. If you need to route to different endpoints within your server, you would handle that internally within your server's logic based on the request's content or headers.
d
@Shubham Bajaj any ideas?
s
yes, can you share the call id please.
PreviousNext
Powered by