How to tell LLM to pass required parameters in a tool call?

[CodeNameR]

We have a Toolkit where we have defined functions like get_available_slots which has required parameters like start_date, end_date etc.

We are using OpenAI and it is making the tool call without passing the start_date or end_date or other parameters.
Is there a way to define which parameters are required? We are currently getting errors like this one -

Missing required argument [type=missing_argument, input_value=ArgsKwargs((), {'start_da... '2024-08-30T10:00:00'}), input_type=ArgsKwargs]\n For further information visit https://errors.pydantic.dev/2.7/

When using OpenAI directly, we could specify required fields in json. How can we achieve the same with PhiData?
{
"name": "get_delivery_date",
"description": "Get the delivery date for a customer's order. Call this whenever you need to know the delivery date, for example when a customer asks 'Where is my package'",
"parameters": {
"type": "object",
"properties": {
"order_id": {
"type": "string",
"description": "The customer's order ID.",
},
},
"required": ["order_id"],
"additionalProperties": false,
}
}

Thanks,
Ritesh

[CodeNameR]

@ashpreetbedi - Is this something you can help with?

[manthanguptaa]

Hey @CodeNameR, I am sharing an example code of a toolkit that you can take inspiration from to develop your tool. My first guess is you didn't register the function.
Code: https://github.com/phidatahq/phidata/blob/main/phi/tools/slack.py