Policy: Azure > AI Foundry > RAI Policy > Content Filters > Input Filter > Jailbreak

The state required for the Jailbreak content filter on the prompt (input) side of the RAI policy. Azure calls this Prompt Shields for jailbreak (direct) attacks.

Each value checks only the fields it names. Enabled requires the filter to be on; Enabled and blocking requires it to be on and blocking; Disabled requires it to be off, and a filter missing from the policy counts as off. Skip leaves this filter unchecked.

Targets

This policy targets the following resource types:

Azure > AI Foundry > RAI Policy

Primary Policy

This policy is used with the following primary policy:

Azure > AI Foundry > RAI Policy > Content Filters > Input Filter

Controls

Setting this policy configures this control:

Azure > AI Foundry > RAI Policy > Content Filters

Policy Specification

Schema Type	`string`
Default	`Skip`
Valid Values [YAML]	`Skip` `Disabled` `Enabled` `Enabled and blocking`
Examples [YAML]	`Enabled and blocking`

In Your Workspace

Policy Settings by Type report

Developers

Category URI

tmod:@turbot/turbot#/control/categories/security

Policy Type URI

tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak

GraphQL

query policyType(id: "tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak") { … }

query policySettings(filter: "policyTypeId:'tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak'") { … }

query policyValues(filter: "policyTypeId:'tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak'") { … }

CLI

Get Policy Type

turbot graphql policy-type --id "tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak"

Get Policy Settings

turbot graphql policy-settings --filter "policyTypeId:tmod:@turbot/azure-aifoundry#/policy/types/raiPolicyContentFiltersInputFilterJailbreak"