A Service agent tells an owner "I could not find any lab results for Bella," even though the front-desk staff can see three released results for that patient in the same org. Testing confirms the action and topic are configured correctly. What is the most likely root cause?
Select an answer to reveal the explanation.
Short Explanation
If the humans can see it but the agent can't, stop looking at the topic and go look at the agent's own badge. Nine times out of ten it is the Agent User missing read access, not the conversation logic.
Full Explanation
When configuration is otherwise verified correct and the agent still cannot surface records that clearly exist, the most common cause is that the Agent User's permission set lacks object- or field-level access to the record in question, so the action's query legitimately returns nothing for that user, and the agent truthfully reports finding no results. This is a security-model problem, not a conversation-design problem. A too-broad classification description would misroute the utterance to the wrong topic entirely or produce an odd response, not a clean "no results" from a correctly-selected action, so it does not fit these symptoms; the scenario states the topic and action are already confirmed correct. An expired portal login for the owner is irrelevant because the query is executed as the Agent User, not as the owner, so the owner's own session state does not change what the action returns. Whether the action is backed by Flow or Apex is a design choice about how logic is implemented, not a factor in whether a user has been granted read access to a record; either backing would fail identically without the right permission set entry.