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

# Understanding Your Chatbot's System Prompt

> Understand what belongs in instructions versus hotel knowledge

Your **System Prompt** tells Lobbychat how to communicate with guests. It sets the chatbot's role, tone, response style, boundaries and what to do when information is uncertain.

## System Prompt versus Knowledge Base

| System Prompt: how to respond | Knowledge Base: what to know |
| - | - |
| Use a warm, professional tone | Breakfast hours and menus |
| Keep simple answers concise | Room amenities and configurations |
| Do not promise actions the chatbot cannot perform | Hotel facilities and policies |
| Refer unconfirmed requests to hotel staff | Contact details and current offers |

For example, if a guest receives the wrong breakfast hours, check the website or other training source. If the answer is accurate but unnecessarily long, review the System Prompt.

## What makes a useful prompt?

Start with a clear role, practical tone guidelines, instructions for uncertainty and operational limits. Avoid copying every hotel fact into the prompt. AI instructions improve consistency, but they cannot guarantee error-free answers.

## Where to edit it

Follow [Edit Your Chatbot's System Prompt](https://documentation.lobbychat.ai/manage/system-prompt.md) for the exact dashboard steps. After saving, select **Retrain Chatbot** and test the result.

**Next:** [Write an effective hotel chatbot prompt](https://documentation.lobbychat.ai/prompting/write-effective-prompt.md).


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