Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
3,864 detections
Filters
Last updated
All Time
Detection languages
2,523
525
299
218
131
Contributors
999
706
554
263
188
Categories
1,448
1,199
1,091
792
575
Platforms
39,931
7,261
6,535
3,864
3,642
Products / Services
1,755
1,449
1,120
438
406
MITRE Techniques
1,888
1,391
913
814
726
CVEs
12
5
4
3
2
IDS Classtypes
10
7
6
6
5
IDS Protocols
37
2
1
This rule detects the use of browser-based cloud shells (AWS CloudShell, Azure Cloud Shell) as an entry point for post-exploitation activities, specifically tracking when these sessions are followed by object storage operations (S3 PutObject/GetObject or equivalent Azure Blob operations) within the same session context. This pattern is indicative of attackers using ephemeral cloud shells to stage stolen data or download offensive tooling, thereby bypassing host-based EDR monitoring.
This rule detects the use of browser-based cloud shells (AWS CloudShell, Azure Cloud Shell) as an entry point for post-exploitation activities, specifically tracking when these sessions are followed by object storage operations (S3 PutObject/GetObject or equivalent Azure Blob operations) within the same session context. This pattern is indicative of attackers using ephemeral cloud shells to stage stolen data or download offensive tooling, thereby bypassing host-based EDR monitoring.
Detects administrative actions intended to disable, modify, or suppress logging and alerting capabilities within AWS and Azure, specifically targeting services like Amazon GuardDuty, AWS Security Hub, AWS Config, and Microsoft Defender for Cloud. Such actions are frequently observed as a precursor to unauthorized activity to hide malicious actions from detection systems.
Detects the deployment of virtual machines with public IP addresses (in Azure) or compute instances with external IP addresses (in GCP) by a single principal within a 15-minute window. This behavior is indicative of a newly-privileged identity or an compromised account establishing externally-accessible infrastructure for further malicious activity.
Detects potential abuse of Cloud Shell to persist access in cloud environments. For GCP, it monitors the injection of SSH keys into instance or project metadata from a Cloud Shell session. For Azure, it detects a pattern involving Cloud Shell storage provisioning followed by a subsequent VM run-command or SSH public key update, indicating potential unauthorized management of virtual machines.
Detects attempts to disable, degrade, or bypass cloud-native security monitoring services in GCP and Azure. This includes disabling GCP Security Command Center, creating broad mute configurations for security findings, downgrading Azure Defender for Cloud pricing tiers, or deleting Azure security automation workflows. These actions are common indicators of defense evasion during an incident.
Detects unauthorized modifications to trust relationships, including Azure AD cross-tenant access policy updates, federation settings, or multi-tenant application credential changes, and GCP Workload/Workforce Identity Federation pool policy updates, when the external entity is not on the approved allow-list.
Detects instances where a long-dormant Entra ID (Azure AD) service principal is reactivated shortly after having a new credential added or a new app role assigned. This activity is a potential persistence technique where an attacker repurposes a legitimate, high-privileged, but unused service principal to blend in as legacy automation.
Detects instances where an administrative user grants 'AllPrincipals' OAuth consent to a multi-tenant application that is unverified or lacks a trusted publisher. This behavior is indicative of an attacker tricking a global admin into granting broad permissions (e.g., Mail.Read, Files.ReadWrite.All) to a malicious application, bypassing per-user consent restrictions.
Detects the addition of a new client secret or certificate to an existing Microsoft Entra ID (formerly Azure AD) application or service principal. This behavior is often used by adversaries to establish persistence following the compromise of an existing identity with administrative permissions, allowing continued access even if the original compromised credentials are revoked.
Detects changes to federation settings on an Entra ID domain, including modifications to IssuerUri, ActiveLogOnUri, MetadataExchangeUri, or SigningCertificate, as well as transitions between Managed and Federated authentication types. Such changes can facilitate 'Golden SAML' attacks by allowing an adversary to forge authentication tokens for the tenant.
Detects the addition of a new federated domain or a modification to an existing domain's authentication settings to Federated, where the domain was not previously verified within a standard verification window. This technique is often used by adversaries to establish persistence and perform self-issued token forgery (e.g., SAML/JWT) by trusting an attacker-controlled identity provider.
Detects changes to federation settings on an Entra ID domain, including modifications to IssuerUri, ActiveLogOnUri, MetadataExchangeUri, or SigningCertificate, as well as transitions between Managed and Federated authentication types. Such changes can facilitate 'Golden SAML' attacks by allowing an adversary to forge authentication tokens for the tenant.
Detects the addition of a new federated domain or a modification to an existing domain's authentication settings to Federated, where the domain was not previously verified within a standard verification window. This technique is often used by adversaries to establish persistence and perform self-issued token forgery (e.g., SAML/JWT) by trusting an attacker-controlled identity provider.
Detects changes to federation settings on an Entra ID domain, including modifications to IssuerUri, ActiveLogOnUri, MetadataExchangeUri, or SigningCertificate, as well as transitions between Managed and Federated authentication types. Such changes can facilitate 'Golden SAML' attacks by allowing an adversary to forge authentication tokens for the tenant.
Detects changes to federation settings on an Entra ID domain, including modifications to IssuerUri, ActiveLogOnUri, MetadataExchangeUri, or SigningCertificate, as well as transitions between Managed and Federated authentication types. Such changes can facilitate 'Golden SAML' attacks by allowing an adversary to forge authentication tokens for the tenant.
Detects when a user is added as an owner to an Azure AD Application or Service Principal, specifically highlighting instances where the new owner does not currently hold a privileged directory role. This pattern is often used as a persistence mechanism to maintain long-term access to an environment without requiring high-level administrative credentials.
Detects the creation or modification of federated identity credentials within Azure Active Directory (Entra ID). This technique allows external services like GitHub Actions, Kubernetes, or other identity providers to authenticate as an application without stored secrets, bypassing traditional secret-rotation and credential-scanning security controls. The rule identifies potential unauthorized persistence by highlighting external or unrecognized issuers.
Detects the addition of secrets or certificates to service principals by identities that are not currently registered as owners of those applications, or modifications occurring outside of standard business hours. This behavior is often indicative of an adversary with compromised administrative privileges (e.g., Application Administrator) attempting to establish persistent, lateral access without alerting legitimate application owners.
Detects a potential Business Email Compromise (BEC) persistence chain where a user grants an OAuth application mail-read or mail-write permissions, followed by the creation of a malicious inbox rule designed to forward, move, or delete sensitive email threads. This combination enables persistent exfiltration and stealth, bypassing password resets as the OAuth token remains active.
This rule detects non-interactive sign-in events using AzureAD tokens from unmanaged devices. This pattern is indicative of potential session or refresh token hijacking, where an adversary replays a stolen session token from their own environment. While a single event is not definitive for impossible travel, it serves as a high-confidence seed for identifying compromised sessions when combined with other indicators.
Page 26 of 184