Intelli
03/12/2025, 3:44 PMVapi Support Bot
03/12/2025, 3:46 PMresult
field must be a single-line string without line breaks [(source)](https://discord.com/channels/1211482211119796234/1211483291191083018/1342188078437236916)
2. Nested JSON problems - If your result contains nested JSON, it must be properly escaped as a string [(source)](https://discord.com/channels/1211482211119796234/1211483291191083018/1334658296291921920)
3. Synchronous vs. asynchronous handling - The server might prematurely mark synchronous tool calls as failed [(source)](https://discord.com/channels/1211482211119796234/1211483291191083018/1342336047589359618)
Known bugs that might affect you:
- Issues with the SDK not properly relaying tool-calls-result messages [(source)](https://discord.com/channels/1211482211119796234/1211483291191083018/1337284665580453969)
- Differences in handling between models like gpt-3.5-turbo
and gpt-4o-realtime-preview
[(source)](https://discord.com/channels/1211482211119796234/1211483291191083018/1329542939160084624)
Try these troubleshooting steps:
1. Double-check your JSON response format:
json
{
"results": [
{
"toolCallId": "<your-tool-call-id>",
"result": "Success message"
}
]
}
2. Check server logs for additional error messages
3. Verify if your tool is configured as synchronous or asynchronous and ensure proper timeout settings
4. Make sure your server returns HTTP status codes in the 200-299 range
Let me know if any of these suggestions help or if you need more specific guidance! 🙂
Sources:
- [Incorrect Response Format](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1325000684231196775)Vapi Support Bot
03/12/2025, 3:46 PMVapi
03/13/2025, 2:01 AMIntelli
03/13/2025, 2:59 PMVapi Support Bot
03/13/2025, 3:00 PM