Azure.AI.ContentSafety Namespace
Important
Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.
Classes
| Name | Description |
|---|---|
| AcsApproval |
Approval metadata returned by an ACS deny decision. Unified Moderate does not invoke an approval workflow. |
| AcsEvidence |
Evidence or a verification pointer returned by policy evaluation. |
| AcsHarmDetail |
The detection result for one blocklist evaluated as part of a harm result. |
| AcsHarmResult |
The moderation and policy-attribution result for one harm category. The category name is the key in the harmResults map and is not restricted to a predefined set. |
| AcsTransform |
A canonical ACS transform applied by Unified Moderate. |
| AcsVerdict |
The complete Agent Control Specification policy verdict. |
| AddOrUpdateTextBlocklistItemsOptions |
The request to add blocklistItems to a text blocklist. |
| AddOrUpdateTextBlocklistItemsResult |
The response of adding blocklistItems to the text blocklist. |
| AnalyzeImageOptions |
The image analysis request. |
| AnalyzeImageResult |
The image analysis response. |
| AnalyzeTextOptions |
The text analysis request. |
| AnalyzeTextResult |
The text analysis response. |
| AzureAIContentSafetyContext |
Context class which will be filled in by the System.ClientModel.SourceGeneration. For more information https://github.com/Azure/azure-sdk-for-net/blob/main/sdk/core/System.ClientModel/src/docs/ModelReaderWriterContext.md |
| BlocklistClient |
The BlocklistClient. |
| BlocklistClientHostExtensions |
Extension methods to add BlocklistClient to an IHostApplicationBuilder. |
| BlocklistClientSettings |
Represents the settings used to configure a BlocklistClient that can be loaded from an IConfigurationSection. |
| ContentProvenanceClient |
The ContentProvenanceClient. |
| ContentProvenanceClientHostExtensions |
Extension methods to add ContentProvenanceClient to an IHostApplicationBuilder. |
| ContentProvenanceClientSettings |
Represents the settings used to configure a ContentProvenanceClient that can be loaded from an IConfigurationSection. |
| ContentSafetyClient |
The ContentSafetyClient. |
| ContentSafetyClientHostExtensions |
Extension methods to add ContentSafetyClient to an IHostApplicationBuilder. |
| ContentSafetyClientOptions |
ContentSafety client options. |
| ContentSafetyClientSettings |
Represents the settings used to configure a ContentSafetyClient that can be loaded from an IConfigurationSection. |
| ContentSafetyImageData |
The image can be either base64 encoded bytes or a blob URL. You can choose only one of these options. If both are provided, the request will be refused. The maximum image size is 2048 x 2048 pixels and should not exceed 4 MB, while the minimum image size is 50 x 50 pixels. |
| ContentSafetyModelFactory |
ContentSafety model factory. |
| DetectedProvenance |
A Microsoft-issued provenance signal indicating the media was created or modified using AI. |
| DetectProvenanceOptions |
Input describing the media to inspect using Content Provenance Detection. |
| DetectProvenanceResult |
Result of a Content Provenance Detection operation. |
| DetectTextProtectedMaterialOptions |
The request of detecting potential protected material present in the given text. |
| DetectTextProtectedMaterialResult |
The combined detection results of potential protected material. |
| DocumentInjectionAnalysisResult |
The individual analysis result of potential injection attacks in the given documents. |
| ImageCategoriesAnalysis |
Image analysis result. |
| ProvenanceContent |
Source content descriptor for a Content Provenance Detection operation. |
| ProvenanceDetectOperation |
Status and result of an asynchronous Content Provenance Detection operation. |
| RemoveTextBlocklistItemsOptions |
The request to remove blocklistItems from a text blocklist. |
| ShieldPromptOptions |
The request of analyzing potential direct or indirect injection attacks. |
| ShieldPromptResult |
The combined analysis results of potential direct or indirect injection attacks. |
| TextBlocklist |
Text Blocklist. |
| TextBlocklistItem |
Item in a TextBlocklist. |
| TextBlocklistMatch |
The result of blocklist match. |
| TextCategoriesAnalysis |
Text analysis result. |
| TextProtectedMaterialAnalysisResult |
The individual detection result of potential protected material. |
| UnifiedModerateConfig |
A request to evaluate and enforce an applicable policy for one workload event. The service uses the first policy it can resolve from policyId and then from the policy binding for targetResourceId, in that order. The request fails if neither selector resolves a policy. |
| UnifiedModerateContext |
Optional agent, session, and request context supplied to policy evaluation. |
| UnifiedModerateResult |
The enforced result of unified moderation. |
| UserPromptInjectionAnalysisResult |
The individual analysis result of potential injection attacks in the given user prompt. |
Structs
| Name | Description |
|---|---|
| AcsDecision |
The canonical decision returned by Agent Control Specification policy evaluation. |
| AnalyzeImageOutputType |
The type of image analysis output. |
| AnalyzeTextOutputType |
The type of text analysis output. |
| DetectedProvenanceType |
Detected provenance record type. |
| DetectOutcome |
Top-level outcome of a Content Provenance Detection operation. |
| ImageCategory |
The harm category supported in Image content analysis. |
| OperationState |
Enum describing allowed operation states. |
| ProvenanceOperationKind |
Kind of asynchronous Content Provenance Detection operation. |
| TextCategory |
The harm category supported in Text content analysis. |
| UnifiedModerateSource |
The workload event source evaluated by Unified Moderate. |
| UnifiedModerateVerdict |
The caller-facing enforcement result returned by Unified Moderate. |
Enums
| Name | Description |
|---|---|
| ContentSafetyClientOptions.ServiceVersion |
The version of the service to use. |
