diff --git a/packages/orchestration/README.md b/packages/orchestration/README.md
index a5abcec04..1f0355c4c 100644
--- a/packages/orchestration/README.md
+++ b/packages/orchestration/README.md
@@ -369,7 +369,10 @@ import {
   buildAzureContentFilter
 } from '@sap-ai-sdk/orchestration';
 
-const filter = buildAzureContentFilter({ Hate: 2, Violence: 4 });
+const filter = buildAzureContentFilter({
+  Hate: AzureFilterThreshold.ALLOW_SAFE_LOW,
+  Violence: AzureFilterThreshold.ALLOW_SAFE_LOW_MEDIUM
+});
 const orchestrationClient = new OrchestrationClient({
   llm: {
     model_name: 'gpt-4o',
@@ -409,7 +412,7 @@ Therefore, handle errors appropriately to ensure meaningful feedback for both ty
 
 `buildAzureContentFilter()` is a convenience function that creates an Azure content filter configuration based on the provided inputs.
 The Azure content filter supports four categories: `Hate`, `Violence`, `Sexual`, and `SelfHarm`.
-Each category can be configured with severity levels of 0, 2, 4, or 6.
+Each category can be configured with severity levels of `ALLOW_SAFE`, `ALLOW_SAFE_LOW`, `ALLOW_SAFE_LOW_MEDIUM` and `ALLOW_ALL` which correspond to Azure threshold values of 0, 2, 4, or 6 respectively.
 
 ### Data Masking
 
diff --git a/packages/orchestration/src/orchestration-types.ts b/packages/orchestration/src/orchestration-types.ts
index fb00c321f..054c56b11 100644
--- a/packages/orchestration/src/orchestration-types.ts
+++ b/packages/orchestration/src/orchestration-types.ts
@@ -153,24 +153,24 @@ export interface DocumentGroundingServiceConfig {
 }
 
 /**
- * A descriptive type for AzureThreshold input.
+ * Filter configuration for Azure Content Safety.
  */
 // eslint-disable-next-line @typescript-eslint/consistent-type-definitions
 export type AzureContentSafety = {
   /**
-   * Filter configuration for Hate.
+   * The filter category for hate content.
    */
   Hate?: AzureFilterThreshold;
   /**
-   * Filter configuration for SelfHarm.
+   * The filter category for self-harm content.
    */
   SelfHarm?: AzureFilterThreshold;
   /**
-   * Filter configuration for Sexual.
+   * The filter category for sexual content.
    */
   Sexual?: AzureFilterThreshold;
   /**
-   * Filter configuration for Violence.
+   * The filter category for violence content.
    */
   Violence?: AzureFilterThreshold;
 };
@@ -179,12 +179,12 @@ export type AzureContentSafety = {
  * A descriptive constant for Azure content safety filter.
  */
 export enum AzureFilterThreshold {
-  /** Allows only safe content. */
+  /** Only safe content is allowed. */
   ALLOW_SAFE = 0,
-  /** Allows safe content and low-risk content. */
+  /** Safe and low-risk content is allowed. */
   ALLOW_SAFE_LOW = 2,
-  /** Allows safe, low-risk, and medium-risk content. */
+  /** Safe, low-risk, and medium-risk content is allowed. */
   ALLOW_SAFE_LOW_MEDIUM = 4,
-  /** Allows all content. */
+  /** All content is allowed. */
   ALLOW_ALL = 6
 }