AWS Security Hub Adds Azure Integration for Cross-Cloud Attack-Path Prioritization

Summary
AWS Security Hub’s Azure connector brings Azure resources and alerts into its exposure engine, correlating ranked findings with AWS to help teams prioritize federated-identity attack paths in one console.
Key points
- The Azure connector records and evaluates Azure resource configurations and streams Microsoft Defender for Cloud alerts and Entra ID audit logs through Azure Event Hubs.
- Security Hub correlates reachability, effective permissions, misconfigurations, vulnerabilities, and other signals into ranked exposure findings for Azure and AWS.
- The walkthrough illustrates potential exposures including an Entra administrator without MFA, an internet-reachable Azure VM with subscription-admin rights, and sensitive data in an AWS S3 bucket.
- The scenario describes configuration-based risks and a possible attack path; it does not report an actual compromise.
- Because Security Hub produces the findings, the top-level cloud provider is AWS; use resources.provider to filter for Azure resources.
- A healthy connector status does not guarantee that resources are being recorded; confirm that Azure Activity Logs reach the event hub and create a remediation task if existing subscriptions lack the required diagnostic setting.
- Security Hub does not raise a finding for a role trusting the organization’s own federated identity provider, so analysts must verify that cross-cloud link in AWS IAM.
Article Details
- Defense Focus
- Correlate cross-cloud identity, configuration, reachability, and permission exposures into ranked attack paths for unified triage.
- Detection Methods
- Correlate exposure findings across cloud environments to identify attack paths involving shared federated identities.
- Use attack-path analysis that combines resource reachability, effective permissions, misconfigurations, and sensitive-data findings.
- Filter findings by the evaluated resource's cloud provider to scope results to Azure resources.
- Verify both connector health and resource-recording activity; a healthy connector does not confirm that resources are being recorded.
- Data Sources
- Azure resource configurations and network security group rules
- Microsoft Defender for Cloud alerts
- Entra ID audit logs and identity authorization policies
- Azure Activity Logs exported to an event hub
- Cloud IAM roles, role assignments, and federated SAML trust configuration
- Vulnerability scanning findings for virtual machines, function apps, and container registry images
- Sensitive-data findings and storage configuration, including bucket encryption, versioning, Object Lock, and MFA delete status
- Platforms
- Multicloud environment spanning AWS and Azure
- Defensive Actions
- Enable the integration across all relevant subscriptions and regions, including the global region scope needed for identity controls.
- Confirm that subscription diagnostic settings export Activity Logs to the connector's event hub; create a remediation task if existing subscriptions are not covered.
- Check that resources are actually being recorded, not just that the connector reports healthy status.
- Review high-severity findings together and investigate the federated trust connecting the identity tenant to the cloud account.
- Address exposed identity and resource conditions, including missing MFA, excessive permissions, permissive inbound network rules, and unprotected sensitive storage.
- For virtual-machine vulnerability scanning, ensure each target VM has a system-assigned managed identity.
MITRE ATT&CK
T1078.004 · Cloud AccountsThe scenario maps use of a compromised cloud identity to Azure access and to assuming a role through the federated trust into AWS.T1190 · Exploit Public-Facing ApplicationThe article maps the modeled foothold to an internet-reachable Azure VM with permissive inbound network rules and an attached identity with administrative rights; exploitation is not reported as an observed event.T1530 · Data from Cloud StorageThe scenario identifies sensitive data in an S3 bucket as reachable through the federated role; access to the data is a modeled risk, not a reported incident.T1537 · Transfer Data to Cloud AccountThe article maps a potential data objective involving the cross-cloud path to this technique, but does not report an actual data transfer.T1550 · Use Alternate Authentication MaterialThe article maps the federated identity trust between the Entra ID tenant and an AWS SAML identity provider to use of alternate authentication material.T1566 · PhishingThe article maps phishing to the modeled initial compromise of an Entra identity; it does not describe a specific phishing message or campaign.
Vendors
Products
Amazon GuardDutyYour Azure resources then flow into the same correlation engine that analyzes signals from AWS Security Hub Cloud Security Posture Management (CSPM), Amazon Inspector, Amazon Macie, and Amazon GuardDuty.Amazon InspectorYour Azure resources then flow into the same correlation engine that analyzes signals from AWS Security Hub Cloud Security Posture Management (CSPM), Amazon Inspector, Amazon Macie, and Amazon GuardDuty.Amazon S3As shown in Figure 2, a role trusts that provider and grants read access to every bucket in Amazon Simple Storage Service (Amazon S3).AWS Security HubAWS Security Hub addresses prioritization and triage with exposure findings.Azure Event Hubsconfigurations, and streams Microsoft Defender for Cloud alerts and Entra ID audit logs through an Azure Event Hubs namespace.MacieYour Azure resources then flow into the same correlation engine that analyzes signals from AWS Security Hub Cloud Security Posture Management (CSPM), Amazon Inspector, Amazon Macie, and Amazon GuardDuty.Microsoft AzureThat analysis now extends to Microsoft Azure.Microsoft Defender for CloudEntra ID and Microsoft Defender for CloudMicrosoft Entra IDConsider an organization whose Microsoft Entra ID tenant is federated into an AWS account, so identities in the tenant can assume an AWS role.
