# What is Guardrail?

A rule that prevents an AI from producing harmful or off-topic output. Guardrails can be built into the model or added via your system prompt.

## In real work

Add a guardrail: never suggest deleting files without confirmation. Always warn the user before any destructive action.

## Related terms

- [System Prompt](https://prompthausapp.com/learn/system-prompt.md)
- [Few-Shot Example](https://prompthausapp.com/learn/few-shot-example.md)
- [Zero-Shot](https://prompthausapp.com/learn/zero-shot.md)
- [Chain of Thought](https://prompthausapp.com/learn/chain-of-thought.md)
- [Persona](https://prompthausapp.com/learn/persona.md)

## How is this different from Googling it?

Google gives a definition. Prompthaus pairs each term with how it shows up in standups, PRs, prompts, and design reviews—then quizzes you until the word sticks. One term a day on iPhone.

- Category: Prompting
- HTML: https://prompthausapp.com/learn/guardrail
- App Store: https://apps.apple.com/us/app/prompthaus-ai-vocabulary/id6776314407