Security Hub CSPM controls for Amazon Bedrock AgentCore
These AWS Security Hub CSPM controls evaluate the Amazon Bedrock AgentCore service and resources. The controls might not be available in all AWS Regions. For more information, see Availability of controls by Region.
[BedrockAgentCore.1] Bedrock AgentCore runtimes should be configured with VPC network mode
Category: Protect > Secure access management > Resource not publicly accessible
Severity: High
Resource type:
AWS::BedrockAgentCore::Runtime
AWS Config rule: bedrockagentcore-runtime-private-network-required
Schedule type: Change triggered
Parameters: None
This control checks whether an Amazon Bedrock AgentCore runtime is configured with VPC network mode. The control fails if the runtime has its network mode set to PUBLIC.
Using public network mode for Amazon Bedrock AgentCore runtimes exposes the runtime directly to the internet, increasing the attack surface and risk of unauthorized access. Configuring runtimes with VPC network mode ensures that runtime traffic is confined within your private network, enabling you to apply network-level security controls such as security groups, network ACLs, and VPC flow logs.
Remediation
To remediate this finding, update the non-compliant Bedrock AgentCore runtime and configure it with VPC network mode. For instructions, see Configure Amazon Bedrock AgentCore Runtime and tools for VPC in the Amazon Bedrock AgentCore Developer Guide.
[BedrockAgentCore.2] Bedrock AgentCore Gateways should require authorization for inbound requests
Category: Protect > Secure access management
Severity: High
Resource type:
AWS::BedrockAgentCore::Gateway
AWS Config rule: bedrockagentcore-gateway-authorizer-enabled
Schedule type: Change triggered
Parameters: None
This control checks whether an Amazon Bedrock AgentCore Gateway requires authorization for inbound requests. The control fails if the Bedrock AgentCore Gateway doesn't have inbound authorization set up.
Configuring authentication on Amazon Bedrock AgentCore gateways ensures that only authorized clients can send requests to your AI agents. Without an authorizer, any entity with network access to the gateway endpoint can invoke your agents, potentially leading to unauthorized data access, resource abuse, or unexpected costs. Inbound authorization validates users who attempt to access targets through your AgentCore gateway.
Remediation
To set up inbound authorization for an Amazon Bedrock AgentCore Gateway, see Set up inbound authorization for your gateway in the Amazon Bedrock AgentCore Developer Guide.
