doc: Usage of Nemo-Guardrails in OpenAI Agent?

[ole2412]

Please also confirm the following

• I have searched the main issue tracker of NeMo Guardrails repository and believe that this is not a duplicate

Issue Kind

Missing documentation

Existing Link

Documentation not found

Description

Hi,
thank you for your work in the guardrails area.
We would like to use a custom guardrail together with a LLM that has access to multiple (database related) tools.
How can we use a nemo guardrail with a LLM that has tool access?

With the OpenAI Agent we were expecting similar to this:

agent = Agent(
    name="Chatbot",
    instructions="You are a helpful assistants. Use the tools to answer the users request.",
    tools=[list_tables, read_query, describe_table],
    output_guardrails=[nemo_custom_guardrail],
    output_type=MessageOutput,
)

Thank you for any help!