SageMaker AI MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AWS_REGION | No | AWS region to use (default: us-east-1) | us-east-1 |
| AWS_PROFILE | No | AWS CLI profile to use for credentials | |
| SAGEMAKER_EXECUTION_ROLE_ARN | Yes | ARN of the SageMaker execution role |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_endpoints_sagemakerA | List all SageMaker Endpoints |
| list_endpoint_configs_sagemakerA | List all SageMaker Endpoint Configurations |
| describe_endpoint_sagemakerB | Describe a SageMaker Endpoint |
| describe_endpoint_config_sagemakerC | Describe a SageMaker Endpoint Configuration |
| delete_endpoint_sagemakerA | Delete a SageMaker Endpoint |
| delete_endpoint_config_sagemakerB | Delete a SageMaker Endpoint Configuration |
| list_training_jobs_sagemakerB | List SageMaker Training Jobs |
| list_processing_jobs_sagemakerB | List SageMaker Processing Jobs |
| list_transform_jobs_sagemakerB | List SageMaker Transform Jobs |
| list_inference_recommendations_jobs_sagemakerA | List all SageMaker Inference Recommender Jobs |
| list_inference_recommendations_job_steps_sagemakerA | List steps for a SageMaker Inference Recommender Job |
| describe_training_job_sagemakerA | Describe a SageMaker Training Job |
| describe_processing_job_sagemakerA | Describe a SageMaker Processing Job |
| describe_transform_job_sagemakerA | Describe a SageMaker Transform Job |
| describe_inference_recommendations_job_sagemakerC | Describe a SageMaker Inference Recommender Job |
| stop_training_job_sagemakerB | Stop a SageMaker Training Job |
| stop_processing_job_sagemakerB | Stop a SageMaker Processing Job |
| stop_transform_job_sagemakerB | Stop a SageMaker Transform Job |
| stop_inference_recommendations_job_sagemakerB | Stop a SageMaker Inference Recommender Job |
| list_pipelines_sagemakerA | List SageMaker Pipelines |
| list_pipeline_executions_sagemakerA | List all Pipeline Executions for a SageMaker Pipeline |
| list_pipeline_execution_steps_sagemakerA | List all Pipeline Execution Steps for a SageMaker Pipeline Execution |
| list_pipeline_parameters_for_execution_sagemakerB | List Pipeline Parameters for a SageMaker Pipeline Execution |
| describe_pipeline_sagemakerC | Describe a SageMaker Pipeline |
| describe_pipeline_definition_for_execution_sagemakerC | Describe Pipeline Definition for a SageMaker Pipeline Execution |
| describe_pipeline_execution_sagemakerD | Describe a SageMaker Pipeline Execution |
| start_pipeline_execution_sagemakerC | Start a SageMaker Pipeline Execution |
| stop_pipeline_execution_sagemakerB | Stop a SageMaker Pipeline Execution |
| delete_pipeline_sagemakerB | Delete a SageMaker Pipeline |
| list_user_profiles_sagemakerA | List all SageMaker User Profiles |
| list_spaces_sagemakerA | List all SageMaker Spaces |
| list_mlflow_tracking_servers_sagemakerA | List all Managed MLflow Tracking Servers in SageMaker |
| create_mlflow_tracking_server_sagemakerC | Create a Managed MLflow Tracking Server in SageMaker |
| create_presigned_url_for_mlflow_tracking_server_sagemakerB | Create a presigned URL for a Managed MLflow Tracking Server in SageMaker |
| describe_mlflow_tracking_server_sagemakerB | Describe a Managed MLflow Tracking Server in SageMaker |
| start_mlflow_tracking_server_sagemakerB | Start a Managed MLflow Tracking Server in SageMaker |
| stop_mlflow_tracking_server_sagemakerB | Stop a Managed MLflow Tracking Server in SageMaker |
| delete_mlflow_tracking_server_sagemakerB | Delete a Managed MLflow Tracking Server in SageMaker |
| list_domains_sagemakerA | List all SageMaker Domains |
| create_presigned_url_for_domain_sagemakerB | Create a presigned URL for a SageMaker Domain |
| describe_domain_sagemakerC | Describe a SageMaker Domain |
| delete_domain_sagemakerB | Delete a SageMaker Domain |
| list_models_sagemakerA | List all SageMaker Models |
| describe_model_sagemakerB | Describe a SageMaker Model |
| delete_model_sagemakerB | Delete a SageMaker Model |
| list_model_cards_sagemakerA | List all SageMaker Model Cards |
| list_model_card_export_jobs_sagemakerA | List Model Card Export Jobs for a SageMaker Model Card |
| list_model_card_versions_sagemakerA | List all versions of a SageMaker Model Card |
| describe_model_card_sagemakerB | Describe a SageMaker Model Card |
| delete_model_card_sagemakerB | Delete a SageMaker Model Card |
| list_apps_sagemakerA | List all SageMaker Apps |
| create_app_sagemakerC | Create a SageMaker App |
| create_presigned_notebook_instance_url_sagemakerA | Create a presigned URL for a SageMaker Notebook Instance |
| describe_app_sagemakerC | Describe a SageMaker App |
| describe_app_image_config_sagemakerC | Describe a SageMaker App Image Config |
| delete_app_sagemakerB | Delete a SageMaker App |
| delete_app_image_config_sagemakerB | Delete a SageMaker App Image Config |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 57 tools
Each tool targets a specific resource-action pair, and even closely related resources like pipelines and pipeline executions are clearly distinguished by their names and descriptions. There is no meaningful overlap that would cause selection ambiguity.
All tools follow a consistent <verb>_<resource>_sagemaker pattern, using snake_case throughout. Verb choice (list, describe, create, delete, start, stop) is uniform, and the resource portion is descriptive and predictable.
With 57 tools, the server is far beyond the typical well-scoped range and exceeds even the 'too many' threshold. The sheer number is overwhelming and suggests an extreme mismatch for an MCP server, even considering the breadth of SageMaker.
The tool set covers many resources but lacks fundamental lifecycle operations such as create or update for core entities like training jobs, models, endpoints, and pipelines. Several listed resources have no describe or delete counterparts, leaving significant gaps that hinder agent workflows.