---
title: SystemLanguageModel.Guardrails
framework: foundationmodels
role: symbol
role_heading: Structure
path: foundationmodels/systemlanguagemodel/guardrails
---

# SystemLanguageModel.Guardrails

Guardrails flag sensitive content from model input and output.

## Declaration

```swift
struct Guardrails
```

## Mentioned in

Improving the safety of generative model output

## Topics

### Getting the guardrail types

- [default](foundationmodels/systemlanguagemodel/guardrails/default.md)
- [permissiveContentTransformations](foundationmodels/systemlanguagemodel/guardrails/permissivecontenttransformations.md)

### Handling guardrail errors

- [LanguageModelError.guardrailViolation(_:)](foundationmodels/languagemodelerror/guardrailviolation(_:).md)

## Relationships

### Conforms To

- [Sendable](swift/sendable.md)
- [SendableMetatype](swift/sendablemetatype.md)

## See Also

### Creating a model for a use case

- [init(useCase:guardrails:)](foundationmodels/systemlanguagemodel/init(usecase:guardrails:).md)
- [SystemLanguageModel.UseCase](foundationmodels/systemlanguagemodel/usecase.md)