CfnAIGuardrailProps
- class aws_cdk.aws_wisdom.CfnAIGuardrailProps(*, assistant_id, blocked_input_messaging, blocked_outputs_messaging, content_policy_config=None, contextual_grounding_policy_config=None, description=None, name=None, sensitive_information_policy_config=None, tags=None, topic_policy_config=None, word_policy_config=None)
Bases:
objectProperties for defining a
CfnAIGuardrail.- Parameters:
assistant_id (
str) – The identifier of the Amazon Q in Connect assistant. Can be either the ID or the ARN. URLs cannot contain the ARN.blocked_input_messaging (
str) – The message to return when the AI Guardrail blocks a prompt.blocked_outputs_messaging (
str) – The message to return when the AI Guardrail blocks a model response.content_policy_config (
Union[IResolvable,AIGuardrailContentPolicyConfigProperty,Dict[str,Any],None]) – Contains details about how to handle harmful content.contextual_grounding_policy_config (
Union[IResolvable,AIGuardrailContextualGroundingPolicyConfigProperty,Dict[str,Any],None]) – The policy configuration details for the AI Guardrail’s contextual grounding policy.description (
Optional[str]) – A description of the AI Guardrail.name (
Optional[str]) – The name of the AI Guardrail.sensitive_information_policy_config (
Union[IResolvable,AIGuardrailSensitiveInformationPolicyConfigProperty,Dict[str,Any],None]) – Contains details about PII entities and regular expressions to configure for the AI Guardrail.tags (
Optional[Mapping[str,str]]) – The tags used to organize, track, or control access for this resource.topic_policy_config (
Union[IResolvable,AIGuardrailTopicPolicyConfigProperty,Dict[str,Any],None]) – Contains details about topics that the AI Guardrail should identify and deny.word_policy_config (
Union[IResolvable,AIGuardrailWordPolicyConfigProperty,Dict[str,Any],None]) – Contains details about the word policy to configured for the AI Guardrail.
- See:
http://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-wisdom-aiguardrail.html
- ExampleMetadata:
fixture=_generated
Example:
# The code below shows an example of how to instantiate this type. # The values are placeholders you should change. from aws_cdk import aws_wisdom as wisdom cfn_aIGuardrail_props = wisdom.CfnAIGuardrailProps( assistant_id="assistantId", blocked_input_messaging="blockedInputMessaging", blocked_outputs_messaging="blockedOutputsMessaging", # the properties below are optional content_policy_config=wisdom.CfnAIGuardrail.AIGuardrailContentPolicyConfigProperty( filters_config=[wisdom.CfnAIGuardrail.GuardrailContentFilterConfigProperty( input_strength="inputStrength", output_strength="outputStrength", type="type" )] ), contextual_grounding_policy_config=wisdom.CfnAIGuardrail.AIGuardrailContextualGroundingPolicyConfigProperty( filters_config=[wisdom.CfnAIGuardrail.GuardrailContextualGroundingFilterConfigProperty( threshold=123, type="type" )] ), description="description", name="name", sensitive_information_policy_config=wisdom.CfnAIGuardrail.AIGuardrailSensitiveInformationPolicyConfigProperty( pii_entities_config=[wisdom.CfnAIGuardrail.GuardrailPiiEntityConfigProperty( action="action", type="type" )], regexes_config=[wisdom.CfnAIGuardrail.GuardrailRegexConfigProperty( action="action", name="name", pattern="pattern", # the properties below are optional description="description" )] ), tags={ "tags_key": "tags" }, topic_policy_config=wisdom.CfnAIGuardrail.AIGuardrailTopicPolicyConfigProperty( topics_config=[wisdom.CfnAIGuardrail.GuardrailTopicConfigProperty( definition="definition", name="name", type="type", # the properties below are optional examples=["examples"] )] ), word_policy_config=wisdom.CfnAIGuardrail.AIGuardrailWordPolicyConfigProperty( managed_word_lists_config=[wisdom.CfnAIGuardrail.GuardrailManagedWordsConfigProperty( type="type" )], words_config=[wisdom.CfnAIGuardrail.GuardrailWordConfigProperty( text="text" )] ) )
Attributes
- assistant_id
The identifier of the Amazon Q in Connect assistant.
Can be either the ID or the ARN. URLs cannot contain the ARN.
- blocked_input_messaging
The message to return when the AI Guardrail blocks a prompt.
- blocked_outputs_messaging
The message to return when the AI Guardrail blocks a model response.
- content_policy_config
Contains details about how to handle harmful content.
- contextual_grounding_policy_config
The policy configuration details for the AI Guardrail’s contextual grounding policy.
- description
A description of the AI Guardrail.
- name
The name of the AI Guardrail.
- sensitive_information_policy_config
Contains details about PII entities and regular expressions to configure for the AI Guardrail.
- tags
The tags used to organize, track, or control access for this resource.
- topic_policy_config
Contains details about topics that the AI Guardrail should identify and deny.
- word_policy_config
Contains details about the word policy to configured for the AI Guardrail.