Keeping prompts genuine (acceptable use)

What prompts are for

Prompts are the real questions we ask AI assistants on your behalf to measure how your brand shows up. They work best as genuine questions a buyer or researcher would actually ask: about your brand, your market, your competitors, or a problem your product solves. See What prompts are and why they matter.

What to avoid

To keep results meaningful and the service healthy for everyone, don't use prompts to:

• Manipulate the assistant: for example trying to make it ignore its instructions, reveal hidden information, or act as something else.
• Run up usage: instructing the assistant to search many times, repeat itself, or produce huge amounts of text.
• Ask for harmful or disallowed content.

Off-topic filler (questions with nothing to do with your brand or market) won't help either: it doesn't measure your visibility.

What happens if a prompt looks like misuse

LumaRank automatically checks prompts for these kinds of misuse. A prompt that looks unsafe is blocked and you'll be asked to rephrase it, and it won't be sent until you do. Repeated attempts can pause your account for review (If your account is paused for review).

If you think a block was a mistake

Our checks are deliberately cautious, so a genuine question is very unlikely to be flagged. If one is, reword it and try again. If your account has been paused and you believe it's a mistake, use the Appeal button to let us know.