----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Create Assistant]() * [Tools Calling]() * [Create Phone Call]() * [Custom LLM]() -----------------------------

### Answer If you're experiencing issues when creating or implementing tools, such as with Make.com or any HTTP-request-based tools in Vapi, it could be related to the setup or configuration of your tool calling structure. Make sure your payload and server configuration are correctly set up following the guidelines in the documentation. You can review and adapt the tool structure as described in the Tools Calling documentation. Here, you can learn how to set up functions, manage asynchronous behavior, and configure server responses. This might help you address any issues with tool implementation [(1)]() . For specific guidance on setting up a tool that interacts with services like Make.com, ensure your payload is properly adapted to meet the functional requirements and that your server is correctly responding to requests from Vapi. Here’s a general example of how a tool configuration might look:

json
{
  "type": "function",
  "messages": [
    {
      "type": "request-start",
      "content": "Starting request..."
    },
    {
      "type": "request-complete",
      "content": "Request complete."
    }
  ],
  "function": {
    "name": "fetch_data",
    "parameters": {},
    "description": "Fetches data from an external API."
  },
  "async": false,
  "server": {
    "url": "https://your-external-api.com/endpoint"
  }
}

If the issues persist, it might be beneficial to check if there are any errors in the server logs or if adjustments are needed in the message formats or API endpoints used within your tools . For further details on troubleshooting and optimizing tool configurations, refer to the Tools Calling section of the Vapi documentation .

Got it. Tagging @User for additional help.