We’re running into an issue where VAPI is triggering the end-of-call report multiple times into our n8n workflow. This is causing duplicate executions, sometimes looping 10–15 times for the same call. We need a way to restrict the end-of-call report so it only triggers once per call. Right now, it keeps retrying even after the first successful trigger. Can you confirm how we can prevent these retries or configure VAPI to only send one report per call ID?

To prevent duplicate processing in your n8n workflow, implement a deduplication check using the

call.id

from the end-of-call report payload. Store processed call IDs (e.g., in a database or n8n's built-in storage) and skip execution if the same ID is received again.

js
if (processedCallIds.includes($json["message"]["call"]["id"])) {
  return; // Already processed, skip
}
// ...process the event and add call.id to processedCallIds

This ensures each call is handled only once, regardless of how many times Vapi retries the webhook. Reference: The webhook payload always includes a unique call ID you can use for this check ([documentation](https://docs.vapi.ai/server-url/events)). Source: - [Server events](https://docs.vapi.ai/server-url/events)

Hi Fiwa, To set up server URLs for receiving and responding to Vapi events, you can follow these steps: 1. **Determine where to set your server URLs**: - **Account-wide**: Set a default server URL in your Vapi dashboard. - **Phone Number**: Attach server URLs to specific phone numbers via the phone number API. - **Assistant**: Configure server URLs for specific assistants either from the dashboard or through the API. - **Function Calls**: Assign server URLs to specific function calls within assistants. You can find more details on setting server URLs [here](https://docs.vapi.ai/server-url/setting-server-urls). 2. **Develop Locally**: - Use a tool like ngrok to create a secure tunnel from your local server to a public URL. This is essential for testing and receiving server events during development. - Start by running your local API, then use

ngrok

to expose your server over the internet. The ngrok URL will be your server URL for Vapi during development. More information on developing locally with Vapi can be found [here](https://docs.vapi.ai/server-url/developing-locally). 3. **Testing with Vapi CLI**: - Use the

vapi listen

command to forward webhook events to your local server. Remember to configure your Vapi webhook URLs to point to the tunnel URL provided by ngrok or any other tunneling service you use. Detailed instructions for local webhook testing setup can be found [here](https://docs.vapi.ai/cli/webhook). By following these steps, you should be able to effectively set up and test server URLs for Vapi.