SanitizeUserPrompt policy

Stay organized with collections Save and categorize content based on your preferences.

This page applies to Apigee and Apigee hybrid.

View Apigee Edge documentation.

Overview

The SanitizeUserPrompt policy protects AI applications from harmful content by sanitizing user prompts sent to generative AI models. The policy uses Model Armor to evaluate user prompts for harmful content, enabling native protection for AI workloads using Apigee. Model Armor is a Google Cloud service offering AI security and safety measures to mitigate the risks associated with large language models (LLMs).

This policy is used in conjunction with the SanitizeModelResponse policy.

This policy is an Extensible policy and use of this policy might have cost or utilization implications, depending on your Apigee license. For information on policy types and usage implications, see Policy types.

Before you begin

Before you use the SanitizeUserPrompt policy, you must complete the following tasks:

• Create a Model Armor template. This step must be completed before you create a SanitizeUserPrompt policy.
• Set the API endpoint for the Model Armor service.
• Create a SanitizeModelResponse policy. This task must be completed before you deploy the SanitizeUserPrompt policy.

Required roles

To get the permissions that you need to apply and use the SanitizeUserPrompt policy, ask your administrator to grant you the following IAM roles on the service account you use to deploy Apigee proxies:

• Model Armor User (roles/modelarmor.user)
• Model Armor Viewer (roles/modelarmor.viewer)

For more information about granting roles, see Manage access to projects, folders, and organizations.

You might also be able to get the required permissions through custom roles or other predefined roles.

Enable APIs

Enable the Model Armor API.

Enable the API

<SanitizeUserPrompt> element

Defines a SanitizeUserPrompt policy.

The <SanitizeUserPrompt> element uses the following syntax:

<SanitizeUserPrompt async="false" continueOnError="false" enabled="true" name="sanitize-text">
  <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
  <DisplayName>Sanitize-Text-sample</DisplayName>
  <ModelArmor>
    <TemplateName>projects/{project}/locations/{location}/templates/{template-name}</TemplateName>
  </ModelArmor>
  <UserPromptSource>{jsonPath('$.contents[-1].parts[-1].text',request.content,true)}</UserPromptSource>
</SanitizeUserPrompt>

<SanitizeUserPrompt async="false" continueOnError="false" enabled="true" name="sanitize-text">
  <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
  <DisplayName>Sanitize-Text-sample</DisplayName>
  <ModelArmor>
    <TemplateName>projects/{project}/locations/{location}/templates/{template-name}</TemplateName>
  </ModelArmor>
  <UserPromptSource>{jsonPath('$.contents[-1].parts[-1].text',request.content,true)}</UserPromptSource>
</SanitizeUserPrompt>

This element has the following attributes that are common to all policies:

The following table provides a high level description of the child elements of <SanitizeUserPrompt>:

{jsonpath('$.input.prompt.text',request.content,false)}

Example

This section provides an example using the <SanitizeUserPrompt>.

This example uses all default values for model detection and prompt extraction:

<SanitizeUserPrompt async="false" continueOnError="false" enabled="true" name="sanitize-text">
  <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
  <DisplayName>Sanitize-Text-sample</DisplayName>
  <ModelArmor>
    <TemplateName>projects/{project}/locations/{location}/templates/{template-name}</TemplateName>
  </ModelArmor>
</SanitizeUserPrompt>

Child element reference

This section describes the child elements of <SanitizeUserPrompt>.

<DisplayName>

Use in addition to the name attribute to label the policy in the management UI proxy editor with a different, more natural-sounding name.

The <DisplayName> element is common to all policies.

The <DisplayName> element uses the following syntax:

<PolicyElement>
  <DisplayName>POLICY_DISPLAY_NAME</DisplayName>
  ...
</PolicyElement>

<PolicyElement>
  <DisplayName>My Validation Policy</DisplayName>
</PolicyElement>

The <DisplayName> element has no attributes or child elements.

<IgnoreUnresolvedVariables>

Determines whether processing stops when a variable is unresolved. Set to true to ignore unresolved variables and continue processing.

<ModelArmor>

Contains the required information for specifying the Model Armor template.

The <ModelArmor> element uses the following syntax:

<ModelArmor>
  <TemplateName>projects/{project}/locations/{location}/templates/{template-name}</TemplateName>
</ModelArmor>

<ModelArmor>
  <TemplateName>projects/{organization.name}/locations/{system.region.name}/templates/{id}</TemplateName>
</ModelArmor>

The following table provides a high-level description of the child elements of <ModelArmor>.

projects/{organization.name}/locations/{system.region.name}/templates/{id}

<UserPromptSource>

The location of the payload for the user prompt text to be extracted. This field supports Apigee message template syntax, including the use of variables or JSON Path functions. For example:

{jsonpath('$.input.prompt.text',request.content,false)}

Flow variables

Flow variables can be used to configure dynamic runtime behavior for policies and flows, based on HTTP headers or message content, or the context available in the Flow. For more information about flow variables, see Flow variables reference.

The policy can set these read-only variables during execution.

Error reference

This section describes the fault codes and error messages that are returned and fault variables that are set by Apigee specific to the SanitizeUserPrompt policy. This information is important to know if you are developing fault rules to handle faults. To learn more, see What you need to know about policy errors and Handling faults.

Runtime errors

These errors can occur when the policy executes.

Deployment errors

These errors can occur when you deploy a proxy containing this policy.

Fault variables

These variables are set when this policy triggers an error at runtime. For more information, see What you need to know about policy errors.

Example error response

{
  "fault": {
    "faultstring": "SanitizeUserPrompt[sanitize-prompt]: unable to resolve variable [variable_name]",
    "detail": {
      "errorcode": "steps.sanitizeuserprompt.UnresolvedVariable"
    }
  }
}

Example fault rule

<FaultRule name="SanitizeUserPrompt Faults">
    <Step>
        <Name>SUP-CustomSetVariableErrorResponse</Name>
        <Condition>(fault.name = "SetVariableFailed")</Condition>
    </Step>
    <Condition>(sanitizeuserprompt.failed = true)</Condition>
</FaultRule>

Model Armor template error code and error messages

The Model Armor template supports overriding error codes and error messages thrown by the SanitizeUserPrompt policy's API requests. If the user sets the overrides, then the Model Armor error codes and error messages will populate the flow variables.

For example, a response with Model Armor error codes and messages could look like the following:

{
  "sanitizationResult": {
    "filterMatchState": "MATCH_FOUND",
    "filterResults": {...},
    "sanitizationMetadata": {
      "errorCode": "890",
      "errorMessage": "get out"
      },
    "invocationResult": "SUCCESS"
  }
}

Schemas

Each policy type is defined by an XML schema (.xsd). For reference, policy schemas are available on GitHub.