AWS sagemaker medium security documentation change

Service: sagemaker · 2026-06-28 · Security-related medium

File: sagemaker/latest/dg/model-registry-deploy.md

Summary

Added code examples for SageMaker Python SDK v3 and v2 (legacy) showing how to deploy model versions, create model objects, and configure cross-account policies for ECR, S3, and KMS resources.

Security assessment

The cross-account policy examples (ECR, S3, KMS) demonstrate security configurations for resource access control. The KMS grant specifically shows security-sensitive operations like 'Decrypt' and 'GenerateDataKey' being delegated to another account.

Diff

diff --git a/sagemaker/latest/dg/model-registry-deploy.md b/sagemaker/latest/dg/model-registry-deploy.md
index 23d4d6b44..44cd0cf72 100644
--- a//sagemaker/latest/dg/model-registry-deploy.md
+++ b//sagemaker/latest/dg/model-registry-deploy.md
@@ -31,0 +32,24 @@ To deploy a model version using the [Amazon SageMaker Python SDK](https://sagema
+SageMaker Python SDK v3
+    
+    
+    
+    from sagemaker.serve import ModelBuilder
+    
+    model_package_arn = 'arn:aws:sagemaker:us-east-2:12345678901:model-package/modeltest/1'
+    
+    # In V3, deploy a model package through ModelBuilder
+    model_builder = ModelBuilder(
+        model=model_package_arn,
+        role_arn=role,
+        sagemaker_session=sagemaker_session,
+        instance_type='ml.m5.xlarge'
+    )
+    model_builder.build()
+    endpoint = model_builder.deploy(
+        initial_instance_count=1,
+        instance_type='ml.m5.xlarge'
+    )
+
+SageMaker Python SDK v2 (Legacy)
+    
+    
@@ -49,0 +74,3 @@ Create a model object from the model version by calling the [create_model](https
+SageMaker Python SDK v3
+    
+    
@@ -60,0 +88,14 @@ Create a model object from the model version by calling the [create_model](https
+SageMaker Python SDK v2 (Legacy)
+    
+    
+        import time
+    import os
+    from sagemaker import get_execution_role, session
+    import boto3
+    
+    region = boto3.Session().region_name
+    
+    role = get_execution_role()
+    
+    sm_client = boto3.client('sagemaker', region_name=region)
+
@@ -75,0 +117,3 @@ Create a model object from the model version by calling the [create_model](https
+SageMaker Python SDK v3
+    
+    
@@ -83,0 +128,14 @@ Create a model object from the model version by calling the [create_model](https
+SageMaker Python SDK v2 (Legacy)
+    
+    
+        import time
+    import os
+    from sagemaker import get_execution_role, session
+    import boto3
+    
+    region = boto3.Session().region_name
+    
+    role = get_execution_role()
+    
+    sm_client = boto3.client('sagemaker', region_name=region)
+
@@ -113,0 +172,3 @@ The following example creates cross-account policies for all three of these reso
+SageMaker Python SDK v3
+    
+    
@@ -218,0 +280,79 @@ The following example creates cross-account policies for all three of these reso
+SageMaker Python SDK v2 (Legacy)
+    
+    
+    
+    import json
+    
+    # The Model Registry account id of the Model Group 
+    model_registry_account = "111111111111"
+    
+    # The model training account id where training happens
+    model_training_account = "222222222222"
+    
+    # 1. Create a policy for access to the ECR repository 
+    # in the model training account for the Model Registry account Model Group
+    ecr_repository_policy = {"Version": "2012-10-17",
+        "Statement": [{"Sid": "AddPerm",
+            "Effect": "Allow",
+            "Principal": {
+              "AWS": f"arn:aws:iam::{model_registry_account}:root"
+            },
+            "Action": [
+              "ecr:BatchGetImage",
+              "ecr:Describe*"
+            ]
+        }]
+    }
+    
+    # Convert the ECR policy from JSON dict to string
+    ecr_repository_policy = json.dumps(ecr_repository_policy)
+    
+    # Set the new ECR policy
+    ecr = boto3.client('ecr')
+    response = ecr.set_repository_policy(
+        registryId = model_training_account,
+        repositoryName = "decision-trees-sample",
+        policyText = ecr_repository_policy
+    )
+    
+    # 2. Create a policy in the model training account for access to the S3 bucket 
+    # where the model is present in the Model Registry account Model Group
+    bucket_policy = {"Version": "2012-10-17",
+        "Statement": [{"Sid": "AddPerm",
+            "Effect": "Allow",
+            "Principal": {"AWS": f"arn:aws:iam::{model_registry_account}:root"
+            },
+            "Action": [
+              "s3:GetObject",
+              "s3:GetBucketAcl",
+              "s3:GetObjectAcl"
+            ],
+            "Resource": [
+              "arn:aws:s3:::{bucket}/*",
+    	  "Resource: arn:aws:s3:::{bucket}"
+            ]
+        }]
+    }
+    
+    # Convert the S3 policy from JSON dict to string
+    bucket_policy = json.dumps(bucket_policy)
+    
+    # Set the new bucket policy
+    s3 = boto3.client("s3")
+    response = s3.put_bucket_policy(
+        Bucket = bucket,
+        Policy = bucket_policy)
+    
+    # 3. Create the KMS grant for the key used during training for encryption
+    # in the model training account to the Model Registry account Model Group
+    client = boto3.client("kms")
+    
+    response = client.create_grant(
+        GranteePrincipal=model_registry_account,
+        KeyId=kms_key_id
+        Operations=[
+            "Decrypt",
+            "GenerateDataKey",
+        ],
+    )
+