I’m having a difficult time getting the AI Assistant to follow my output format instructions. It always adds additional “helpful” information that I do not want and have multiple statements in my prompt instructing it to follow my output format specifically. I’ve engage in conversations to see if the assistant could offer suggestions. This yielded some additional text I’ve added the prompts, but then it consistently ignores these. These are not large prompts. The Assistant indicated it is following its system prompts to supply additional information. Is there a solution to this?
Hi @kentrad
Welcome to the Quick Suite community!
Q assistant in Quick Suite has system-level behavior that governs how it structures responses, adding explanatory context, caveats, and supplementary narrative. Your custom instructions and prompt text cannot override this system behavior, which is why the assistant itself told you it’s following its system prompts.
If you are using Data Stories: The story prompt can give you some narrative control (for example, create a concise summary with only 4 bullet points), but the assistant may still append its own framing text. You can edit the generated story afterward to remove any content you don’t want.
If you are using Q&A (Data Q&A): The response format is largely fixed by the system. Topic‑level and field‑level descriptions can steer what data is returned, but they do not control how the assistant presents that data in its narrative.
I’ve marked this as a feature request to promote visibility to the support team.