Class Guardrail (0.3.0)

Guardrail(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Guardrail contains a list of checks and balances to keep the agents safe and secure.

This message has oneof_ fields (mutually exclusive fields). For each oneof, at most one member field can be set at the same time. Setting any member of the oneof automatically clears all other members.

.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields

Attributes
Name	Description
content_filter	google.cloud.ces_v1beta.types.Guardrail.ContentFilter Optional. Guardrail that bans certain content from being used in the conversation. This field is a member of oneof_ guardrail_type.
llm_prompt_security	google.cloud.ces_v1beta.types.Guardrail.LlmPromptSecurity Optional. Guardrail that blocks the conversation if the prompt is considered unsafe based on the LLM classification. This field is a member of oneof_ guardrail_type.
llm_policy	google.cloud.ces_v1beta.types.Guardrail.LlmPolicy Optional. Guardrail that blocks the conversation if the LLM response is considered violating the policy based on the LLM classification. This field is a member of oneof_ guardrail_type.
model_safety	google.cloud.ces_v1beta.types.Guardrail.ModelSafety Optional. Guardrail that blocks the conversation if the LLM response is considered unsafe based on the model safety settings. This field is a member of oneof_ guardrail_type.
code_callback	google.cloud.ces_v1beta.types.Guardrail.CodeCallback Optional. Guardrail that potentially blocks the conversation based on the result of the callback execution. This field is a member of oneof_ guardrail_type.
name	str Identifier. The unique identifier of the guardrail. Format: projects/{project}/locations/{location}/apps/{app}/guardrails/{guardrail}
display_name	str Required. Display name of the guardrail.
description	str Optional. Description of the guardrail.
enabled	bool Optional. Whether the guardrail is enabled.
action	google.cloud.ces_v1beta.types.TriggerAction Optional. Action to take when the guardrail is triggered.
create_time	google.protobuf.timestamp_pb2.Timestamp Output only. Timestamp when the guardrail was created.
update_time	google.protobuf.timestamp_pb2.Timestamp Output only. Timestamp when the guardrail was last updated.
etag	str Etag used to ensure the object hasn't changed during a read-modify-write operation. If the etag is empty, the update will overwrite any concurrent changes.

Classes

CodeCallback

CodeCallback(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Guardrail that blocks the conversation based on the code callbacks provided.

ContentFilter

ContentFilter(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Guardrail that bans certain content from being used in the conversation.

LlmPolicy

LlmPolicy(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Guardrail that blocks the conversation if the LLM response is considered violating the policy based on the LLM classification.

LlmPromptSecurity

LlmPromptSecurity(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Guardrail that blocks the conversation if the input is considered unsafe based on the LLM classification.

This message has oneof_ fields (mutually exclusive fields). For each oneof, at most one member field can be set at the same time. Setting any member of the oneof automatically clears all other members.

.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields

ModelSafety

ModelSafety(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Model safety settings overrides. When this is set, it will override the default settings and trigger the guardrail if the response is considered unsafe.