What Be Nice mode does
When you submit a prompt in Be Nice mode, S.I.M.P. analyzes it for the patterns most likely to cause AI hallucination — vague scope, missing context, implicit assumptions, and underspecified output requirements. Each problem is surfaced as an issue with a short type label and a one-sentence description explaining what’s wrong and why it matters. After listing the issues, S.I.M.P. rewrites your prompt into a sanitized version with the problems resolved. A changes log explains exactly what was edited and why each change reduces hallucination risk. The summary gives you a plain-language overview of the prompt’s weaknesses and what was done to address them.When to use Be Nice mode
- Starting out. Be Nice mode is the default — it’s a good first pass for any prompt, regardless of your experience level.
- Professional or collaborative contexts. If you’re sharing results with a team, a client, or a manager, the constructive framing is easier to present.
- Learning prompt engineering. The issue descriptions explain why something is a problem, not just that it is one. Over time, you’ll start catching the same patterns before you even submit.
What the feedback looks like
Each issue in Be Nice mode follows a calm, explanatory tone. For example:- Type:
vague_scope— Description: “The prompt doesn’t define the boundaries of the task, which may cause the AI to make assumptions about what’s in or out of scope.” - Type:
missing_context— Description: “No background information is provided, so the AI may invent details about the subject to fill the gap.”