https://vapi.ai/ logo
Join Discord
Powered by
Issue with 'tools'
# support
l
.
v
Hi @Leo, I'm an AI assistant for Vapi. I'll analyze available sources and get back to you shortly.
----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Tools Calling]() * [Function Calling]() * [Server Events]() * [Create Assistant]() -----------------------------
### Answer The structure you've provided for using 'tools' in the Vapi platform seems well-formed according to the documentation on configuring tools within your Vapi assistant [(1)]() . It's important to ensure that the server URL is correctly set up and can handle requests from Vapi as expected. If your assistant fails to call your functions when using 'tools', but succeeds with 'functions', there might be specific configurations or issues related to how your server handles these tool calls or possibly how the 'tools' are set up within your assistant's configuration. One potential issue could be related to how messages are managed or the asynchronous behavior of the tool calls. If using 'functions' leads to the assistant saying 'just a sec/just a moment', it indicates that the assistant is waiting for a response from your server. When you switch to 'tools', and it doesn't work as expected, it might be due to how these tool calls are being processed or triggered. Make sure that: 1. Your server is correctly handling incoming requests from Vapi's tool calls. The server needs to respond correctly to the payload sent by Vapi. 2. The 'async' property is set according to your needs. If set to false, the assistant will wait for the function to complete before continuing the conversation. This might not be suitable for all use cases, especially if you need the conversation to proceed without interruptions. 3. Ensure that all required fields in the tool's configuration are correctly filled out, including the 'server' object with a valid URL and any necessary 'function' details like 'name', 'parameters', and 'description'. If after checking these configurations the issue persists, it might be helpful to look into the server's logs to see if there are any errors or issues when processing the requests
from Vapi. Additionally, reviewing the assistant's settings in the Vapi dashboard and ensuring that the 'tools' configuration matches your application's needs might provide more insight. Without direct access to more detailed error logs or feedback messages from your server or the Vapi platform, it's challenging to provide a more specific solution. However, ensuring that your server is correctly set up to handle requests from Vapi and reviewing the assistant's configuration for any discrepancies in how 'tools' are defined might help address the issue.
Thanks for the follow-up, @Leo. I'll get back to you soon.
----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Tools Calling]() * [Server Events]() * [Call Ended Reason]() * [Function Calling]() -----------------------------
### Answer Sorry, I wasn't able to find any information about your question.
Got it. Tagging @User for additional help.
s
Yeah, it is supposed to do like that you just need to extract correct information from it and then send it back!
l
so basically what the'tools' sends to my endpoint is totally different from the one sent by 'functions'?
s
Yup.
l
is there a way to just use 'functions' but just stop it from saying 'just a moment' every time
s
Nope. The only way through which you can fix it by using Tools Calling.
l
@Sahil I've now managed to trigger the tools, but it's still saying 'hold on a sec' when the assistant try to trigger the tool call, what do I need to do to disable the mention of 'hold on a sec' or 'just a sec' etc?
s
did you remove function from server and client message and replaced with tools calling?
l
I have both under server and client message
so just remove 'function-call' for both serverMessages and clientMessages?
issue persist after removing @Sahil
### Also, is the response format correct like above Sahil?
s
Reply in the text format in the result parameter. Also, can you send me a call_id?
So, that I can check once
l
can you clarify what you mean by -> 'Reply in the text format in the result parameter.' ?
5923377e-2249-4ef9-a803-ba063a2d37ce @Sahil can you take a look at this call id?
s
Copy code
json

{
    "results": [
        {
            "toolCallId": "call_BZzoTQQfR7etw3Bb0ph5FeX5",
            "result": "tomorrow at 8 o'clock in the morning, EST"
        }
    ]
}
}
This is what I mean
l
noted, can you please remove this message Sahil? thanks
s
Done.
l
thanks Sahil
s
Is it working now?
l
nope, reverted back to using 'functions' for now, so far didn't manage to get it working yet
8 Views
PreviousNext
Powered by