BrowserStack AI Evals
EvaluationGuardrails

Guardrails

Enforce guardrails around your model calls from the SDK.

Guardrails

Guardrails are safety and policy rules that act on your model calls — redacting sensitive text, returning a safe response, throwing an error, or re-asking the model. You author, validate, and promote them in the platform; the SDK enforces the live rules in your application.

For what guardrails are, how to author them, and how they behave, see the Guardrails guide in the platform documentation.

This section covers the SDK surface:

Manual Enforcement

Run guardrails explicitly over your own content with check / enforce.

Manage Guardrails as Code

Author, fetch, list, update, and archive guardrails from code.

Evaluators

Create evaluator lists and run inline evaluations across TypeScript, Python, and Java SDKs.

Manual Enforcement

Apply guardrails explicitly over your own content with check (verdict, never throws) and enforce (throws on block).

On this page

Guardrails