Amazon SageMaker AI MCP
Official AWS Labs MCP for SageMaker HyperPod cluster management: deployment, node management, lifecycle operations on EKS-orchestrated or Slurm-orchestrated clusters. Read-only by default; --allow-write and --allow-sensitive-data-access flags gate destructive operations.
“Official AWS Labs MCP for SageMaker HyperPod cluster management: deployment, node management, lifecycle operations on EKS-orchestrated or Slurm-orchestrated clusters. Read-only by default; --allow-write and --allow-sensitive-data-access flags gate destructive operations. 2 commits on the server path in the last 30 days, the lowest cadence in this batch; Tier 2 placement reflects the cadence.”
INSTALL THIS SERVER
{
"mcpServers": {
"awslabs.sagemaker-ai-mcp-server": {
"command": "uvx",
"args": [
"awslabs.sagemaker-ai-mcp-server@latest"
],
"env": {
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}
{
"mcpServers": {
"awslabs.sagemaker-ai-mcp-server": {
"command": "uvx",
"args": [
"awslabs.sagemaker-ai-mcp-server@latest"
],
"env": {
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}
{
"mcpServers": {
"awslabs.sagemaker-ai-mcp-server": {
"command": "uvx",
"args": [
"awslabs.sagemaker-ai-mcp-server@latest"
],
"env": {
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}
{
"mcpServers": {
"awslabs.sagemaker-ai-mcp-server": {
"command": "uvx",
"args": [
"awslabs.sagemaker-ai-mcp-server@latest"
],
"env": {
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}
{
"mcpServers": {
"awslabs.sagemaker-ai-mcp-server": {
"command": "uvx",
"args": [
"awslabs.sagemaker-ai-mcp-server@latest"
],
"env": {
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}
6 SELECTED TOOLS
OUR ASSESSMENT
The SageMaker AI MCP focuses on HyperPod, the dedicated training cluster surface inside SageMaker. Operators provision clusters, manage nodes, and run lifecycle operations (start, stop, scale) on clusters orchestrated by Amazon EKS or Slurm. Default operating mode is read-only; the --allow-write flag opens cluster mutations and the --allow-sensitive-data-access flag opens privileged read paths. Stack Protection ensures CloudFormation stacks created by the tool can only be modified by the same tool.
- Apache-2.0 licence.
- Official AWS Labs maintenance with 8,924 parent-repo stars.
- Read-only default with explicit --allow-write opt-in.
- Stack Protection guards against cross-tool CloudFormation stack mutation.
- Configurable AWS profile and region via standard environment variables.
- 2 commits on the server path in the last 30 days, lowest cadence in this batch.
- HyperPod is the only SageMaker surface covered; full SageMaker control-plane operations live elsewhere.
- File system access is broad (the README documents this as a security consideration): the server can read and write any path the host user can.
- Designed for local stdio use; HTTP transport carries additional risk per the README.
The server defaults to read-only mode. Open --allow-write only for sessions where cluster mutation is intended. Open --allow-sensitive-data-access only when access to sensitive HyperPod data is required. Run under a dedicated AWS profile with HyperPod-scoped IAM permissions. The host file system is accessible to the server; scope the agent filesystem accordingly. The README warns against network exposure of this server.
ML platform teams running SageMaker HyperPod for distributed training jobs; AWS shops where training infrastructure runs inside SageMaker for compliance and IAM unification with the rest of the AWS workload; operators who want a guarded MCP entry point for HyperPod cluster operations (read-only by default; explicit flags for write paths).
TECHNICAL DETAILS
RECORDED ADOPTION METRICS
Saved directory snapshots. Record last checked MAY 1, 2026. The review date above does not refresh these figures.
// Recorded note8,924 stars on the awslabs/mcp parent monorepo. 2 commits on the SageMaker AI server path in the last 30 days.
// Recorded noteFifth-ranked in ai-ml. Tier 2 cadence; Tier 1 vendor signal.
SOURCES & RECORDED CHECKS
These are the directory listings and check details held in this record. Client compatibility, directory listings and installation testing are separate signals. Recorded client names describe the scope of a check; they do not establish a security assessment or a recurring test schedule.
The same server, 4 different lenses. We reconcile these signals into our editorial score, which is why our number sometimes diverges from a directory-aggregate star count.
| Source | Their rating | Their star count | Their downloads | Last synced |
|---|---|---|---|---|
| AutomationSwitch This page | 3.9editorial | 8,924 | — | MAY 1, 2026 |
| PulseMCP | — unrated | unavailable | unavailable | MAY 1, 2026 |
| MCP.so | — unrated | unavailable | unavailable | MAY 1, 2026 |
| Glama | — unrated | unavailable | unavailable | MAY 1, 2026 |
| Official MCP Registry | — unrated | unavailable | unavailable | MAY 1, 2026 |
// Counts are directory-reported; we don't adjust them. Discrepancies usually come from different snapshot times or star-caching.
OTHER AI / ML MCP SERVERS
Cognee MCP
Knowledge graph plus vector memory engine for AI agents, exposed as an MCP server with V2 session-aware memory tools (remember, recall, forget, improve) and classic V1 ingestion pipelines (cognify, codify). Three transports: stdio, SSE, Streamable HTTP. 16,965 GitHub stars, Apache-2.0.
HeyGen Hyperframes
Write HTML, render video. Built for agents. HeyGen's framework that turns HTML templates into video output via MCP. 16,400 GitHub stars and 100 commits on main in the last 30 days.
Codebase Memory MCP
High-performance code intelligence MCP server for AI coding agents. Indexes a codebase into a queryable knowledge graph in milliseconds, with 14 MCP tools spanning structural search, call-chain tracing, impact analysis, dead-code detection, and Cypher queries. Single static C binary, 66 languages via tree-sitter, zero runtime dependencies.
Arize Phoenix MCP
LLM observability platform exposing prompts, projects, traces, spans, sessions, datasets, and experiments through MCP. Published to npm as @arizeai/phoenix-mcp, current 4.0.8 (2026-04-29). 9,496 stars on parent monorepo, Elastic License 2.0.
Qdrant MCP Server
Official Qdrant vector database MCP server. Acts as a semantic memory layer on top of Qdrant: store information with metadata, retrieve via similarity search. Two tools, very small surface area, exceptionally maintained by the Qdrant team. Configurable embedding provider (fastembed default), supports remote and local Qdrant clusters.
Mastra
TypeScript framework for building AI-powered applications and agents, with built-in MCP server support. From the team behind Gatsby. 23,680 GitHub stars and 100 commits on main in the last 30 days.
DISCUSS YOUR
MCP REQUIREMENTS.
Evaluating a server, scoping an internal deployment, or working out whether MCP is the right fit at all. Start the conversation and we will point you at the right piece of the ecosystem.