Assistant says there has been an error when there ...
# support
We have an assistant that when asked to, displays a meeting booking form on the screen for the user, this is called using a function. However, when it does this the assistant always says something along the lines of "This will just take a sec. Apologize for the inconvenience. It seems though there was an issue with setting up the meeting. Let's try this again. Can you please provide your name and the name of your business? And I'll make sure to get a meeting scheduled for you. To learn more about neural voices personal applications." But the meeting booking form still shows up and works. Not sure why the assistant seems to think there is an error
Hi @Jeremy, I'm an AI assistant for Vapi. While a team member reviews this question, I'll look through relevant sources and see if I can provide suggestions.
----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Outbound Sales Example]() * [Set up your Server URL]() * [Create Assistant]() * [Update Assistant]() -----------------------------
### Answer It seems like the assistant might be responding based on the outcome of a function call, indicating an error even when the function (displaying the booking form) executes correctly. This discrepancy could be due to how the function call's result is handled or interpreted within your assistant's logic. To address this, you might want to review the logic surrounding your function call within the assistant's configuration, ensuring that the response handling accurately reflects the actual outcome of the function call. Specifically, checking how the assistant interprets the function call's result to decide what message to convey to the user. If the function is intended to display the meeting booking form and does so successfully, the logic should ensure that a success message, or no error message, is relayed to the user instead of an error message. Additionally, ensure that the server handling the function call returns the expected response to the assistant. If the server response is misinterpreted as an error by the assistant, it could lead to the behavior you're observing. You might want to refer to the documentation on [Function Calling]() [(1)]() and ensure your server's response aligns with what the assistant expects. If this doesn't resolve the issue, it might be beneficial to examine the assistant's configuration for any conditional statements or error handling mechanisms that might inadvertently trigger the display of an error message. For further insights on configuring your assistant, you can refer to the [Create Assistant]() [(2)]() and [Update Assistant]() [(3)](htt
ps:// documentation. This guidance assumes a configuration or logic mismatch, and reviewing both the assistant's response logic and the server-side handling of the function call is recommended.
Most probably the issue is with your function call URL. You are not sending the right response from it. Also, in case that's not the issue then send me your call_id