Skip to main content

Safety and data protection

Guardrails are project-scoped bundles of content and safety rules the gateway enforces inline: on the way in (prompts) and on the way out (responses). Each guardrail is a collection of rules; each rule is a check type scored by a guardrail provider. About guardrails explains that model; the guides below cover the rest.

About guardrails

The model behind guardrails: rules, providers, evaluation order, and failure modes.

Configure guardrails

Create guardrails and rules; validate in monitor mode before enforce.

Configure guardrail providers

Choose regexp, builtin, or tetrate for each check type.

Detect and redact sensitive data

Keep PII and credentials out of prompts and responses.

Detect and block prompt injection

Stop jailbreak patterns before they reach a model.

Protect coding-assistant traffic

Apply stricter guardrails to agent and editor traffic without slowing normal use.

Guardrails on shared data planes

Per-project enforcement on shared capacity, gateway attachment health, and identity alerts.

Protect requests with guardrails

The developer-side view: recognizing a block or a redaction and requesting coverage.