PlatformsTechnical DevelopmentSeptember 24, 2026

AWS AgentCore Gateway AI enables cross-account data queries for enterprises

AWS AgentCore Gateway AI enables cross-account data queries for enterprises

AWS has released a reference architecture for deploying AI agents that query data across multiple AWS accounts using Amazon Bedrock AgentCore Gateway and the Model Context Protocol (MCP). This solution enables enterprises to maintain data sovereignty within individual accounts while providing a unified endpoint for cross-account tool discovery and authorization.

Key Takeaways

  • Amazon Bedrock AgentCore Gateway acts as a central hub for tool discovery and invocation across distributed accounts.
  • Model Context Protocol servers run on AgentCore Runtime to expose local data as structured tools like credit scoring or policy search.
  • Security is managed through AgentCore Identity and Okta, using OAuth 2.0 machine-to-machine credentials for cross-account authentication.
  • Policy in AgentCore uses the Cedar language to enforce fine-grained, per-user authorization at the gateway layer before routing requests.

Why It Matters

This architecture addresses a critical friction point for large-scale streaming and media enterprises: the need for intelligent agents to access siloed data without the security risk of mass replication. By keeping data in its original account and only moving specific query results, AWS provides a scalable path for deploying RAG-based tools across complex organizational structures. This move strengthens the Amazon Bedrock ecosystem against competitors by offering a managed framework for governance and cost attribution at the account level. Industry observers should watch for the adoption of the Model Context Protocol as a standard for interoperability between specialized media microservices and central AI control planes.

Additional Context

The integration of FAST channel workflow orchestration within this gateway framework allows media companies to automate metadata injection and ad-insertion triggers across distributed cloud environments. Furthermore, the focus on production readiness for agentic workflows highlights the broader industry shift toward securing data access for enterprise-grade AI deployments, a trend recently underscored by the launch of the CData Connect AI Gateway to govern enterprise agent data access, which complements new modality-aware cost management tools for managing AI infrastructure.

Companies in this story

In short

AWS has launched the AgentCore Gateway AI, a new architecture enabling AI agents to query data across distributed AWS accounts without centralizing datasets. By utilizing the Model Context Protocol, this system maintains data sovereignty while providing a unified endpoint, helping large-scale enterprises deploy secure, RAG-based tools across complex organizational structures.

FAQ

What is the purpose of the AWS AgentCore Gateway?

The gateway acts as a central hub for tool discovery and invocation, allowing AI agents to query data across multiple AWS accounts without needing to move or centralize datasets.

How does the Model Context Protocol work with AgentCore?

Model Context Protocol servers run on the AgentCore Runtime to expose local data as structured tools, such as policy search or credit scoring, for use by AI agents.

How is security managed in this multi-account architecture?

Security is handled through AgentCore Identity and Okta using OAuth 2.0 credentials, while the Cedar language enforces fine-grained, per-user authorization at the gateway layer.

Why is this architecture important for media enterprises?

It allows intelligent agents to access siloed data without the security risks associated with mass replication, providing a scalable path for governance and cost attribution.


Read full article at aws.amazon.com

Enjoy our coverage?

Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.

Add as preferred source

Related Articles

Get this in your inbox → Subscribe