Reference documentation and code samples for the Google Cloud Ces V1 Client class LlmPromptSecurity.
Guardrail that blocks the conversation if the input is considered unsafe based on the LLM classification.
Generated from protobuf message google.cloud.ces.v1.Guardrail.LlmPromptSecurity
Namespace
Google \ Cloud \ Ces \ V1 \ GuardrailMethods
__construct
Constructor.
| Parameters | |
|---|---|
| Name | Description |
data |
array
Optional. Data for populating the Message object. |
↳ default_settings |
LlmPromptSecurity\DefaultSecuritySettings
Optional. Use the system's predefined default security settings. To select this mode, include an empty 'default_settings' message in the request. The 'default_prompt_template' field within will be populated by the server in the response. |
↳ custom_policy |
LlmPolicy
Optional. Use a user-defined LlmPolicy to configure the security guardrail. |
↳ fail_open |
bool
Optional. Determines the behavior when the guardrail encounters an LLM error. - If true: the guardrail is bypassed. - If false (default): the guardrail triggers/blocks. Note: If a custom policy is provided, this field is ignored in favor of the policy's 'fail_open' configuration. |
getDefaultSettings
Optional. Use the system's predefined default security settings.
To select this mode, include an empty 'default_settings' message in the request. The 'default_prompt_template' field within will be populated by the server in the response.
| Returns | |
|---|---|
| Type | Description |
LlmPromptSecurity\DefaultSecuritySettings|null |
|
hasDefaultSettings
setDefaultSettings
Optional. Use the system's predefined default security settings.
To select this mode, include an empty 'default_settings' message in the request. The 'default_prompt_template' field within will be populated by the server in the response.
| Parameter | |
|---|---|
| Name | Description |
var |
LlmPromptSecurity\DefaultSecuritySettings
|
| Returns | |
|---|---|
| Type | Description |
$this |
|
getCustomPolicy
Optional. Use a user-defined LlmPolicy to configure the security guardrail.
| Returns | |
|---|---|
| Type | Description |
LlmPolicy|null |
|
hasCustomPolicy
setCustomPolicy
Optional. Use a user-defined LlmPolicy to configure the security guardrail.
| Parameter | |
|---|---|
| Name | Description |
var |
LlmPolicy
|
| Returns | |
|---|---|
| Type | Description |
$this |
|
getFailOpen
Optional. Determines the behavior when the guardrail encounters an LLM error.
- If true: the guardrail is bypassed.
- If false (default): the guardrail triggers/blocks. Note: If a custom policy is provided, this field is ignored in favor of the policy's 'fail_open' configuration.
| Returns | |
|---|---|
| Type | Description |
bool |
|
setFailOpen
Optional. Determines the behavior when the guardrail encounters an LLM error.
- If true: the guardrail is bypassed.
- If false (default): the guardrail triggers/blocks. Note: If a custom policy is provided, this field is ignored in favor of the policy's 'fail_open' configuration.
| Parameter | |
|---|---|
| Name | Description |
var |
bool
|
| Returns | |
|---|---|
| Type | Description |
$this |
|
getSecurityConfig
| Returns | |
|---|---|
| Type | Description |
string |
|