Skip to content
worth noting AI agents

AWS architecture for connecting Bedrock agents to private APIs through a centralized network

only one source so far

AWS has released a guide on an architecture pattern that allows Bedrock AgentCore Gateway agents to call private REST APIs and MCP servers with a central inspection layer. The pattern uses Transit Gateway and VPC Lattice, and includes CloudFormation templates and test scripts for AWS Organizations.

AWS has released a guide on an architectural pattern for connecting Bedrock AgentCore Gateway agents with private REST APIs and Model Context Protocol (MCP) servers. The pattern is intended for enterprise environments where a central network team owns all virtual private networks and requires that all traffic between them pass through a central firewall in both directions.

The pattern uses the self-managed mode of the VPC Lattice service with a resource gateway located in the network account. As a result, all traffic between agents and private APIs or MCP servers passes through Transit Gateway and the central inspection layer. The guide defines the division of responsibility: the network team manages the resource gateway and its infrastructure, the team managing the tools configures the load balancer and OAuth, and the team responsible for Bedrock agents sets up the agents themselves and their targets.

AWS tested the pattern in the US East (N. Virginia) region in a lab environment with three member accounts and one management account in AWS Organizations. The publication includes CloudFormation templates, test scripts, validation checks, an address plan, and troubleshooting materials. The guide serves as a reference that companies can adapt to their own infrastructure.

What changed

Why it matters

Companies that deploy Bedrock agents and need them to communicate with private APIs often face a requirement for centralized network inspection. A tested and documented pattern speeds up implementation, eliminates the need for custom development, and simplifies maintenance. It makes it easier for engineers to standardize their approach and reduces security risks for IT teams through a verified design.

Two audiences, two different impacts

What this means

01

For individuals

When implementing agents with private APIs, you can rely on a tested pattern instead of a custom design; ready-made templates and configuration guidelines are available for the individual roles.

What to do Consider deploying this pattern when addressing agent integration with private APIs; review the available CloudFormation templates and examples from the AWS repository.
More practical updates →
02

For a business

Companies can standardize the deployment of Bedrock agents in enterprises with centralized network management; the pattern speeds up implementation and reduces security risks through a verified design.

Processes
What to decide Add this pattern to the catalog of recommended architectures for agents; train devops and network teams on this approach for cases involving a centralized inspection layer.
More business impacts →
AgentCore Gateway Amazon Bedrock Model Context Protocol private API Transit Gateway VPC Lattice

Check the original

Event sources

only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected Connect Amazon Bedrock AgentCore Gateway to private APIs in a hub-and-spoke network with centralized inspection