AWS launches a serverless MCP data lake gateway
Understand this story
The Amazon NASC team built a serverless MCP server on Amazon ECS Fargate as a unified gateway for AI agents to access the enterprise data lake. According to AWS, adding a new dataset requires only a single JSON metadata file, and the service automatically discovers it on the next pipeline sync, with no integration code to write or redeployment needed; the architecture supports any MCP-compatible client, and permissions are enforced uniformly at the MCP server layer. The architecture is available only in AWS Regions that offer both Amazon Bedrock and Amazon Bedrock Knowledge Bases. In testing, end-to-end interaction latency was 15–20 seconds. Scoping retrieval on demand is optional and can be added at any time without modifying metadata files; if no context is specified, the agent retrieves all indexed tables.
AI overview based on source reports · Updated 2 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- AWS Machine Learning BlogServerless MCP server for AI agent data lake access on Amazon ECS Fargate
Amazon's North America Sort Center team built a serverless Model Context Protocol (MCP) server on Amazon ECS Fargate that acts as a centralized, agent-agnostic gateway between AI agents and the enterprise data lake. Adding a dataset requires only a metadata JSON file, which the pipeline indexes into Amazon Bedrock Knowledge Bases, while Amazon Cognito authentication, SELECT-only Athena tools, Bedrock Guardrails, and CloudWatch logging provide centralized controls. The architecture separates the tool layer from the agent layer so any MCP-compatible client can share the same tools and access controls in a multi-tenant environment.
Activity for this story
Not enough continuous observations to show a trend yet.