PRODUCTION SCENARIO
Vantage Health runs a tool-calling agent under Zero Data Retention with store set to false. Between steps the agent forwards only the function_call and function_call_output items, and quality on multi-step cases has dropped compared with the stored version.
What should the agent include in each follow-up request?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: All reasoning items, function call items, and function call outputs since the last user message
With store set to false, reasoning items carry an encrypted_content property, and passing every reasoning item, function call, and function call output since the last user message lets the model continue its reasoning across consecutive tool calls in the most token-efficient way. Summaries are not a substitute, and previous_response_id relies on server-side state that a Zero Data Retention configuration does not keep.
Official Source: https://developers.openai.com/api/docs/guides/reasoning ↗