Azure

This content introduces AI Detection and Response (AIDR) as a new cybersecurity category to address emerging threats from autonomous AI agents, including supply chain attacks and unintended data sharing, highlighting their ability to execute with inherited privileges across endpoints, SaaS, and cloud environments.

Elastic has developed a detection rule that monitors non-allowlisted `wget` activity on Linux hosts using Auditd Manager or Auditbeat, leveraging an Elastic LLM to triage `wget` executions for potential ingress tool transfer, command and control, or data exfiltration attempts to untrusted destinations, generating alerts only for high-confidence positive or suspicious verdicts.

Elastic's LLM-based detection rule identifies suspicious `curl` activity on Linux systems, aiming to detect command and control, data exfiltration, or ingress tool transfer by analyzing command-line parameters and network destinations via Auditd Manager or Auditbeat logs, which, if left unaddressed, could lead to system compromise or data breach.

CrowdStrike highlights a critical identity management gap in AI agent deployments where current OAuth 2.1 tokens and JWT (RFC 9068) lack standardized mechanisms to represent an AI agent's instance identity, the user on whose behalf it acts, and their relationship, hindering fine-grained access controls, audit trails, and detection of out-of-scope actions.

Tracking brief for the Shai-Hulud campaign; individual sightings are folded in as reported.

CrowdStrike's AI security research team has identified 18 new prompt injection techniques, expanding its taxonomy to over 200 methods, which enable adversaries to manipulate AI systems and agents through indirect means like hidden context, delayed triggers, and special token injection, leading to unauthorized actions such as data exfiltration or arbitrary command execution.

Roundup of Microsoft security advisories published in July 2026.

This rule detects the abuse of Azure Virtual Machine Run Command to execute scripts remotely, correlating Azure Activity Log events with endpoint process starts, identifying instances where adversaries use Run Command to run scripts as SYSTEM or root.

This rule detects Linux process executions that access high-value Kubernetes service-account material, kubeconfig or node PKI paths, or common cloud files, potentially indicating credential theft within in-cluster and hybrid environments.

Microsoft disrupted Fox Tempest, a threat actor running a malware-signing-as-a-service (MSaaS) that abuses Microsoft Artifact Signing to generate short-lived code-signing certificates used to sign malware disguised as legitimate software, delivering ransomware and various information stealers to victims across multiple sectors.

Hackers injected credential-stealing malware into newly published versions of the node-ipc npm package in a supply chain attack, collecting cloud credentials, SSH keys, CI/CD secrets, and other sensitive data, exfiltrating it through DNS TXT queries.

Multiple vulnerabilities in Microsoft developer tools and platforms could allow an attacker to achieve arbitrary code execution, data manipulation, privilege escalation, bypassing security measures, information disclosure, and denial of service.

Multiple vulnerabilities in Microsoft Azure and Windows Admin Center allow an attacker to escalate privileges, spoof information, and bypass security measures.

Multiple vulnerabilities exist in Microsoft Azure, specifically affecting azl3 kernel and azl3 krb5, potentially leading to an unspecified security issue.

A compromised version (7.0.4) of the intercom-client npm package was published using a compromised developer account, containing obfuscated JavaScript that executed during installation to harvest and exfiltrate credentials from the environment, as part of the 'Mini Shai-Hulud' supply chain campaign.

Multiple vulnerabilities in Microsoft Azure, Microsoft 365 Copilot, Microsoft Dynamics 365, and Microsoft Power Apps could allow an attacker to escalate privileges, execute arbitrary code, and conduct spoofing attacks.

Adversaries are abusing Entra ID device code authentication using a malicious broker client to bypass MFA and gain unauthorized access to Azure resources by compromising Primary Refresh Tokens (PRTs).

A single user deleting multiple Azure Restore Point Collections in a short time period can indicate a ransomware attack or destructive operation, preventing victim recovery by inhibiting system recovery.

A recently disclosed vulnerability allows attackers to bypass Azure sign-in logs, potentially masking malicious activity within cloud environments.

A suspicious browser activity alert indicates anomalous behavior based on suspicious sign-in activity across multiple tenants from different countries in the same browser, potentially indicating compromised credentials or other malicious activity.

A cloud compute instance was created with a previously unseen image, potentially indicating malicious activity such as unauthorized deployment or image compromise.

An attacker may add an authentication method to a compromised Azure account for persistent access, which can be detected by monitoring changes to authentication methods in Azure audit logs.

Detection of unauthorized access or privilege escalation attempts within Azure environments due to invalid or missing Microsoft Entra Premium P2 or Microsoft Entra ID Governance licenses for Privileged Identity Management (PIM).

Detection of Azure Network Watcher's Packet Capture feature being enabled, potentially indicating malicious network sniffing for credential access and discovery of sensitive data in unencrypted traffic.

Detects Azure service principals authenticating from multiple countries within a short time, indicating potentially compromised credentials being used from different geographic locations.

Adversaries may delete Azure diagnostic settings to evade defenses by hindering detection and analysis, which this detection identifies by monitoring Azure activity logs for successful deletion operations.

An Azure firewall was created, modified, or deleted, potentially indicating malicious activity aimed at impairing network defenses.

Detection of Azure Privileged Identity Management (PIM) elevation approvals or denials, which, if unexpected, may indicate unauthorized privilege escalation or malicious activity within an Azure environment.

Detection of Azure Privileged Identity Management (PIM) roles being activated without requiring multi-factor authentication, potentially leading to unauthorized privilege escalation and persistence.

This alert identifies when an application is deleted within an Azure environment, which could indicate malicious activity or unintended misconfiguration leading to service disruption.

Detection of a user being assigned the 'User Access Administrator' role, which grants the ability to manage all Azure Subscriptions, potentially leading to privilege escalation and unauthorized access.

Detection of an excessive number of Global Administrator accounts assigned within an Azure tenant, indicating potential privilege escalation or compromised accounts.

Detects the creation of a service principal in Azure, which could indicate potential attacker activity for lateral movement or persistence.

Detection of a service principal removal in Azure, potentially indicating malicious activity or an attempt to remove evidence of a compromise.

Detection of assigned but unused privileged roles in Azure's Privileged Identity Management (PIM) service, indicating potential misconfiguration, license overuse, or dormant privileged access that could be exploited.

Detection of a failed attempt to invite an external guest user by an Azure user lacking the necessary permissions, potentially indicating privilege escalation or malicious insider activity.

This rule detects suspicious processes, such as copy utilities or scripting tools, accessing sensitive identity files on Linux systems, including Kubernetes tokens, cloud CLI configurations, and root SSH keys, indicating potential credential theft.

An adversary disables Privileged Identity Management (PIM) alerts in Azure to evade detection and maintain persistent access with escalated privileges.

The rule detects mass deletion of Azure disk snapshots, which could indicate an adversary attempting to inhibit system recovery capabilities, destroy backup evidence, or prepare for a ransomware attack.

This rule detects Linux process executions that access sensitive Kubernetes, cloud, and SSH credential files via common utilities, potentially indicating credential theft.

Detection of frequent role activation in Azure Privileged Identity Management (PIM) by the same user may indicate potential privilege escalation or account compromise.

An adversary may add a user account as an owner for an Azure application in order to grant additional permissions and modify the application's configuration using another account, potentially leading to persistence, credential access, or privilege escalation.

Anomalous addition of credentials to an Entra ID service principal by a user not typically performing this action can indicate potential persistence and privilege escalation by an attacker.

Detection of privileged Azure AD role assignments to users, which can indicate persistence and privilege escalation by threat actors.

Detects the creation of new privileged accounts in Azure environments, potentially indicating initial access, persistence, privilege escalation, or stealth activities by malicious actors.

An attacker elevates their Azure subscription permissions to manage all subscriptions, potentially leading to unauthorized access and control over the environment.

Detection of a user being assigned a built-in administrator role in Azure RBAC, which can be abused for privilege escalation, lateral movement, or persistence.

An adversary may remove an owner from an Azure application or service principal to weaken access controls, persist in the environment, or escalate privileges.

An adversary may add unauthorized credentials to an Azure application, enabling persistent access, evading defenses, and escalating privileges by modifying certificates or secrets.

This brief focuses on detecting malicious inbox manipulation rules set within a user's Azure environment, often indicative of account compromise or insider threats aiming to conceal illicit activities.

Detection of unauthorized guest user invitations within an Azure Active Directory tenant, indicating potential privilege escalation, persistence, or initial access attempts.

An adversary may add a user account as an owner for an Azure service principal to define what an application can do in the Azure AD tenant, potentially leading to persistence and privilege escalation.

Detection of Azure alert suppression rule creation or modification events, which can be used by attackers to disable security alerts and evade detection.

A single user or service principal deleting multiple Azure Storage Accounts within a short time period may indicate malicious activity such as data destruction, service disruption, or a ransomware attack.