AWS sagemaker high security documentation change

Service: sagemaker · 2026-07-10 · Security-related high

File: sagemaker/latest/dg/mlflow-app-setup-prerequisites-iam.md

Summary

Added detailed IAM permissions requirements for Model Registry sync and lifecycle management, including policy examples and lifecycle transition controls

Security assessment

The change adds critical IAM policy configurations required for secure model lifecycle management. It explicitly addresses security controls by demonstrating how to restrict model promotion to production stages using IAM condition keys (sagemaker:ModelLifeCycle/stage and stageStatus), which prevents unauthorized lifecycle transitions. The documentation of required S3 permissions and lineage tracking also helps prevent unintended data access.

Diff

diff --git a/sagemaker/latest/dg/mlflow-app-setup-prerequisites-iam.md b/sagemaker/latest/dg/mlflow-app-setup-prerequisites-iam.md
index 759a7a372..3340b6bc1 100644
--- a//sagemaker/latest/dg/mlflow-app-setup-prerequisites-iam.md
+++ b//sagemaker/latest/dg/mlflow-app-setup-prerequisites-iam.md
@@ -7 +7 @@
-Set up MLflow IAM permissions when creating a new domainCreate necessary IAM service roles in the IAM consoleCreate action-specific authorization controls
+Set up MLflow IAM permissions when creating a new domainCreate necessary IAM service roles in the IAM consoleCreate action-specific authorization controlsIAM permissions for Model Registry sync and lifecycle management
@@ -165,0 +166,107 @@ The following SageMaker AI MLflow actions are supported for authorization access
+## IAM permissions for Model Registry sync and lifecycle management
+
+When you register MLflow models with the SageMaker AI Model Registry, the MLflow App IAM service role requires the following permissions to perform model registration, lifecycle management, and lineage tracking:
+    
+    
+    {
+        "Version": "2012-10-17",
+        "Statement": [
+            {
+                "Effect": "Allow",
+                "Action": [
+                    "sagemaker:CreateModelPackageGroup",
+                    "sagemaker:DescribeModelPackageGroup",
+                    "sagemaker:CreateModelPackage",
+                    "sagemaker:UpdateModelPackage",
+                    "sagemaker:AddTags"
+                ],
+                "Resource": "*"
+            },
+            {
+                "Sid": "S3ArtifactAccess",
+                "Effect": "Allow",
+                "Action": [
+                    "s3:GetObject",
+                    "s3:PutObject",
+                    "s3:ListBucket"
+                ],
+                "Resource": "*"
+            },
+            {
+                "Sid": "LineageTracking",
+                "Effect": "Allow",
+                "Action": [
+                    "sagemaker:CreateAction",
+                    "sagemaker:AddAssociation"
+                ],
+                "Resource": "*"
+            }
+        ]
+    }
+
+The following table describes what each permission enables:
+
+Action | Purpose  
+---|---  
+`sagemaker:CreateModelPackageGroup` | Creates a Model Package Group when you register a new model name.  
+`sagemaker:DescribeModelPackageGroup` | Checks whether a Model Package Group already exists before creating one.  
+`sagemaker:CreateModelPackage` | Creates a Model Package (version) when you register a model version.  
+`sagemaker:UpdateModelPackage` | Updates the Model Package lifecycle stage and status when you set lifecycle aliases.  
+`sagemaker:AddTags` | Tags the Model Package with metadata linking it back to the MLflow model.  
+`s3:GetObject` | Reads inference specification and evaluation artifacts from Amazon S3 during registration.  
+`sagemaker:CreateAction`, `sagemaker:AddAssociation` | Creates lineage associations between the MLflow model and the SageMaker AI Model Package. If missing, lineage is silently skipped.  
+  
+### Restrict lifecycle transitions with IAM condition keys
+
+You can use IAM condition keys to control which lifecycle transitions a role is allowed to perform. The following condition keys are available for the `sagemaker:UpdateModelPackage` action:
+
+  * `sagemaker:ModelLifeCycle/stage` — The lifecycle stage being set. Values: `staging`, `production`.
+
+  * `sagemaker:ModelLifeCycle/stageStatus` — The lifecycle status being set. Values: `pending`, `active`.
+
+
+
+
+**Example: Deny production promotions**
+
+The following policy prevents a role from promoting models to the `production` stage:
+    
+    
+    {
+        "Version": "2012-10-17",
+        "Statement": [
+            {
+                "Effect": "Deny",
+                "Action": "sagemaker:UpdateModelPackage",
+                "Resource": "*",
+                "Condition": {
+                    "StringEquals": {
+                        "sagemaker:ModelLifeCycle/stage": "production"
+                    }
+                }
+            }
+        ]
+    }
+
+**Example: Allow only staging/pending transitions**
+
+The following policy restricts a role to only set models to the `staging` stage with `pending` status:
+    
+    
+    {
+        "Version": "2012-10-17",
+        "Statement": [
+            {
+                "Effect": "Allow",
+                "Action": "sagemaker:UpdateModelPackage",
+                "Resource": "*",
+                "Condition": {
+                    "StringEquals": {
+                        "sagemaker:ModelLifeCycle/stage": "staging",
+                        "sagemaker:ModelLifeCycle/stageStatus": "pending"
+                    }
+                }
+            }
+        ]
+    }
+