> ## Documentation Index
> Fetch the complete documentation index at: https://simpai.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Frequently asked questions about S.I.M.P.

> Answers to the most common questions about S.I.M.P. — how it works, what the risk score means, why prompts get flagged, and how to get the best results.

S.I.M.P. (Sanitized Input Model Prompt) is a free web app that analyzes your AI prompts, scores their hallucination risk on a 1–10 scale, and rewrites them for clarity. The questions below cover the most common things people ask before and after their first use.

<AccordionGroup>
  <Accordion title="Is S.I.M.P. free to use?">
    Yes. S.I.M.P. is completely free — no account, no signup, and no credit card required. Open the app and start analyzing prompts immediately.
  </Accordion>

  <Accordion title="Does S.I.M.P. answer my prompt?">
    No. S.I.M.P. does not answer your prompt. It analyzes and rewrites your prompt to reduce hallucination risk, then hands it off to an AI model of your choice. Use the **ChatGPT**, **Claude**, or **Gemini** buttons to send your sanitized prompt to the model that will actually respond.

    <Note>
      The ChatGPT button opens chatgpt.com, the Claude button opens claude.ai/new, and the Gemini button copies your sanitized prompt to the clipboard so you can paste it yourself.
    </Note>
  </Accordion>

  <Accordion title="What's the difference between Be Nice and Roast Me mode?">
    Both modes apply the same quality fixes to your prompt — they identify the same issues and produce equally effective sanitized prompts. The only difference is tone:

    * **Be Nice** delivers feedback in a constructive, straightforward way.
    * **Roast Me** delivers the same feedback with a savage, comedic edge.

    Pick whichever keeps you engaged. The output quality is identical.
  </Accordion>

  <Accordion title="What does the hallucination risk score mean?">
    The risk score is a number from 1 to 10 that estimates how likely your prompt is to cause an AI model to make things up (hallucinate). A higher score means the prompt has more gaps, ambiguity, or missing constraints that invite the model to fill in the blanks with fabricated information.

    * **1–3 (Low):** Your prompt is clear and well-scoped. Minor adjustments only.
    * **4–6 (Moderate):** Some gaps exist. The sanitized version addresses them.
    * **7–8 (High):** Significant issues detected. Review the issues panel carefully.
    * **9–10 (Critical):** The prompt is very likely to produce unreliable output. Major rewriting applied.
  </Accordion>

  <Accordion title="Why did S.I.M.P. give me a High or Critical score?">
    A High or Critical score usually means your prompt is missing one or more of the following:

    * **Context** — the prompt doesn't explain the situation or background
    * **Scope** — the prompt is too broad or open-ended
    * **Constraints** — there are no limits on length, format, tone, or topic
    * **Specificity** — key terms are vague or undefined

    Open the **Issues** panel to see exactly what S.I.M.P. flagged and review the **What Changed & Why** section to understand how the sanitized version addresses each problem.
  </Accordion>

  <Accordion title="Can I run my prompt through S.I.M.P. more than once?">
    Yes. Click the **Run Again** button to load your sanitized prompt back into the input field. You can then submit it for another round of analysis and refinement. Iterating two or three times on complex prompts often produces noticeably better results.
  </Accordion>

  <Accordion title="Does S.I.M.P. store my prompts?">
    S.I.M.P. uses [Simple Analytics](https://simpleanalytics.com) for usage tracking — a privacy-friendly analytics tool that does not use cookies and does not collect personal data. The app does not display a data retention policy, and we won't speculate about server-side storage. If data handling is important to your use case, avoid submitting sensitive or confidential information.
  </Accordion>

  <Accordion title="Which AI model does S.I.M.P. use?">
    S.I.M.P. uses **Anthropic Claude Sonnet** as its primary model for analyzing and rewriting your prompts. If Claude is unavailable, it automatically falls back to a secondary model. You don't need to configure anything — the fallback happens automatically.
  </Accordion>

  <Accordion title="What do I do after I get my sanitized prompt?">
    Once S.I.M.P. has finished analyzing your prompt, use one of the handoff buttons to take the sanitized version to an AI model:

    * **ChatGPT** — opens chatgpt.com in a new tab
    * **Claude** — opens claude.ai/new in a new tab
    * **Gemini** — copies the sanitized prompt to your clipboard; paste it into Gemini yourself

    You can also copy the sanitized prompt manually and paste it into any other AI tool you prefer.
  </Accordion>

  <Accordion title="Is there a limit to how many times I can use S.I.M.P.?">
    Yes. S.I.M.P. applies rate limiting to keep the service available for everyone. You can send up to **30 requests in a burst**, and the limit refills at **10 requests per minute**. If you hit the limit, you'll see the message "Rate limit exceeded. Try again in a moment." — wait about 60 seconds and you'll be able to continue.
  </Accordion>
</AccordionGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.