Amazon SageMaker AI MCP Server
Deploy and manage SageMaker HyperPod clusters on EKS or Slurm from an AI assistant.
TypeMCP server
CategoryDev & data
Made byAWS
PricingNeeds AWS account
RunsOn your computer
Checked on2026-11-19
What it does
Currently focused on HyperPod: agents can deploy clusters, manage nodes and handle lifecycle operations for ML training infrastructure.
What an agent can do with it
- deploy HyperPod clusters
- manage cluster nodes
- update cluster software
- describe clusters
How to set it up
Runs locally with uvx (Python package from PyPI) or Docker; uses your local AWS credentials/profile and IAM permissions.
Full steps: AWS's documentation.
Sources
Descriptions are written by Agents Lookup unless marked as quoted. Each link opens the original.
- MCP documentation: github.com/awslabs/mcp/tree/main/src/sagemaker-ai-mcp-server (checked 2026-11-19)
- AWS list where it appears: github.com/awslabs/mcp