It really needs a fix from trieve team, you need to include a flag to only allow message of the query instead of the whole chunks retrieval. It makes my call hallucinate and expensive for no fucking reason. I am manually creating Trieve's api and parsing only the answer from it and will serve it as vapi tool now.

You are correct—Trieve currently retrieves whole chunks, and there is no flag to return only the answer to a query. As a workaround, you can create a custom Vapi tool that calls your Trieve API, parses the response, and returns only the answer you need. To do this, update your assistant's configuration to include your new tool, test it in a development environment, and then deploy it to production. Example step:

json
{
  "tools": [
    {
      "name": "trieve-answer-tool",
      "endpoint": "https://your-api/parse-trieve-answer",
      "method": "POST"
    }
  ]
}

This approach lets you control the output and avoid unnecessary chunk retrieval. Sources: - [Trieve Integration chunks retrieval behavior](https://discord.com/channels/1211482211119796234/1334102591470047292) - [vapi tool creation and usage](https://discord.com/channels/1211482211119796234/1333975440003563604)

Hey, Could you please schedule a brief call at your convenience so we can discuss this matter in detail? Kindly use the following link to book a suitable time: .

bro your calendly says you are booked for this month, this issue wont make any sense if i talk about it later.

Hey! To help track down this issue, could you share: - The call ID - When exactly this happened (the timestamp) - What response you expected to get - What response you actually got instead This would really help us figure out what went wrong!