Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
839 detections
Filters
Last updated
All Time
Detection languages
438
190
77
62
40
Contributors
120
114
72
71
68
Categories
464
233
193
152
122
Platforms
288
274
154
88
56
Products / Services
220
169
162
90
76
MITRE Techniques
18,116
15,638
12,587
8,397
6,227
839
CVEs
6
5
5
5
5
IDS Classtypes
4
2
IDS Protocols
1
Detects a successful device-code sign-in followed within 60 minutes by non-interactive replay from a different IP AND a persistence action (inbox rule, app consent, device add) from a non-original IP. Persistence is now mandatory (inner join) instead of optional, and the retired AADSignInEventsBeta table is replaced by EntraIdSignInEvents.
The following analytic detects the addition of a servicePrincipalName (SPN) containing invisible Unicode characters to an Active Directory computer object via Windows Security Event 5136.
This is the key indicator of the KerberLoss attack (CVE-2026-25177), where an attacker with write access to a machine account's SPN attribute injects a collision SPN that contains a Zero Width or other invisible Unicode character (e.g. U+200C Zero Width Non-Joiner).
LDAP's string preparation rules (RFC 4518) cause these characters to be ignored during uniqueness checking, allowing the write to succeed even when an identical SPN already exists on another machine account.
The Kerberos KDC, however, does not apply the same normalisation — it finds two accounts with matching SPNs and returns KDC_ERR_S_PRINCIPAL_UNKNOWN, causing a Kerberos denial-of-service or forcing clients to fall back to NTLM downgrade.
This is the key indicator of the KerberLoss attack (CVE-2026-25177), where an attacker with write access to a machine account's SPN attribute injects a collision SPN that contains a Zero Width or other invisible Unicode character (e.g. U+200C Zero Width Non-Joiner).
LDAP's string preparation rules (RFC 4518) cause these characters to be ignored during uniqueness checking, allowing the write to succeed even when an identical SPN already exists on another machine account.
The Kerberos KDC, however, does not apply the same normalisation — it finds two accounts with matching SPNs and returns KDC_ERR_S_PRINCIPAL_UNKNOWN, causing a Kerberos denial-of-service or forcing clients to fall back to NTLM downgrade.
Detects the creation or modification of system users and groups on NetScaler ADC/Gateway appliances by monitoring Syslog events (e.g., nsconmsg, audit logs). The rule specifically triggers when these actions occur outside of standard business hours or on weekends, indicating potential unauthorized account manipulation or unauthorized persistence mechanisms.
The following analytic detects a user account (non-machine account) adding or removing a servicePrincipalName (SPN) on a computer object in Active Directory, via Windows Security Event 5136.
In normal AD operations, SPN values on computer objects are managed exclusively by the computer account itself (during domain join or name change), by Domain Controllers during replication, or by the SYSTEM/NETWORK SERVICE context — all of which appear as accounts ending in the dollar sign ($) convention.
A named user account writing to the servicePrincipalName attribute of a computer object is anomalous and may indicate an attacker with delegated WriteProperty rights over computer accounts performing SPN manipulation.
In normal AD operations, SPN values on computer objects are managed exclusively by the computer account itself (during domain join or name change), by Domain Controllers during replication, or by the SYSTEM/NETWORK SERVICE context — all of which appear as accounts ending in the dollar sign ($) convention.
A named user account writing to the servicePrincipalName attribute of a computer object is anomalous and may indicate an attacker with delegated WriteProperty rights over computer accounts performing SPN manipulation.
Detects modifications to Microsoft Entra ID (formerly Azure AD) authentication methods policy where a new external authentication method is configured using a non-approved discovery URL. This could indicate an adversary attempting to configure a malicious identity provider to facilitate authentication bypass or unauthorized access.
Detects instances where an identity associated with AI or automated agents (service accounts, service principals, bots, or assistants) is assigned to highly privileged administrative roles or security groups in the directory. This behavior potentially indicates an attempt to escalate privileges using a compromised or improperly configured agent identity.
Detects when a user has either been assigned or removed from an Admin/Privileged role. Providing the analyst with the relevant event information for further investigation.
The following analytic detects a user account modifying its own userPrincipalName (UPN) to match the sAMAccountName of another account via Windows Security Event 4738.
This is the setup step of the ResetNightmare attack (CVE-2026-27912), where an attacker with WriteProperty rights on their own UPN attribute spoofs their identity to a target account.
The KDC then resolves a ptype=10 (NT-ENTERPRISE) Kerberos pre-authentication request against the spoofed UPN, issuing a kadmin/changepw TGT that can be used to change the target account's password via kpasswd (port 464).
Event 4738 is generated when a user account attribute is changed. This analytic filters to events where the SubjectUserSid equals the TargetSid (self-modification), the new UPN value is not a standard UPN (no @ sign), and the value is not a Windows placeholder.
A non-UPN value set on one's own account is anomalous and has no legitimate administrative use case.
This is the setup step of the ResetNightmare attack (CVE-2026-27912), where an attacker with WriteProperty rights on their own UPN attribute spoofs their identity to a target account.
The KDC then resolves a ptype=10 (NT-ENTERPRISE) Kerberos pre-authentication request against the spoofed UPN, issuing a kadmin/changepw TGT that can be used to change the target account's password via kpasswd (port 464).
Event 4738 is generated when a user account attribute is changed. This analytic filters to events where the SubjectUserSid equals the TargetSid (self-modification), the new UPN value is not a standard UPN (no @ sign), and the value is not a Windows placeholder.
A non-UPN value set on one's own account is anomalous and has no legitimate administrative use case.
The following analytic identifies when the SID of a privileged user is added to the SID History attribute of another user. It leverages Windows Security Event Codes 4742 and 4738, combined with identity lookups, to detect this activity. This behavior is significant as it may indicate an attempt to abuse SID history for unauthorized access across multiple domains. If confirmed malicious, this activity could allow an attacker to escalate privileges or maintain persistent access within the environment, posing a significant security risk.
Detects modifications to Microsoft Entra ID (formerly Azure AD) authentication policies, specifically the registration, enablement, or modification of External Authentication Methods (EAMs). This activity can indicate an attempt to add a rogue authentication provider for persistent unauthorized access, bypassing standard MFA or conditional access policies.
The following analytic detects the assignment of privileged Azure Active Directory roles to a user. It leverages Azure AD audit logs, specifically monitoring the "Add member to role" operation. This activity is significant as adversaries may assign privileged roles to compromised accounts to maintain persistence within the Azure AD environment. If confirmed malicious, this could allow attackers to escalate privileges, access sensitive information, and maintain long-term control over the Azure AD infrastructure.
The following analytic detects potential privilege escalation threats in Azure Active Directory (AD). This detection is important because it identifies instances where privileged roles that hold elevated permissions are assigned to service principals. This prevents unauthorized access or malicious activities, which occur when these non-human entities access Azure resources to exploit them. False positives might occur since administrators can legitimately assign privileged roles to service principals. This detection leverages the O365 Universal Audit Log data source.
Detects the creation of new administrator accounts or super_admin profiles on network devices such as firewalls from Fortinet, Cisco, Meraki, WatchGuard, and Sophos. This rule monitors log messages indicating administrative account provisioning, which is a potential indicator of unauthorized account manipulation or persistence attempts.
Detects instances where a Service Principal or Enterprise Application is assigned a highly privileged Entra ID directory role (e.g., Global Administrator, Privileged Role Administrator). This behavior is often associated with adversary attempts to establish persistent access that survives remediation of human-managed identities.
This rule detects when a new password credential (client secret) is added to an existing Azure AD Service Principal or application. This activity is a common method for adversaries to establish persistence in a cloud environment that survives password resets or MFA enforcement on the original administrative account.
Detects when a user registers a new multi-factor authentication (MFA) method, such as an authenticator app, phone number, or FIDO key, within 30 minutes of a sign-in event flagged as anomalous (e.g., impossible travel, new ASN, or risky sign-in score). This pattern is indicative of an adversary establishing persistence by registering their own MFA device to maintain access to a compromised account after a session hijack.
Detects MFA or security information registration, updates, or deletions within Entra ID AuditLogs that occur within two hours of a sign-in event flagged as risky, impossible travel, or involving a new device. This sequence is a common indicator of post-compromise identity persistence.
Detects account modification activity (e.g., password resets or MFA registration changes) followed within 24 hours by a sign-in event for the same user account that is flagged as risky or exhibits anomalous device/location characteristics, indicative of social engineering campaigns like Scattered Spider.
Detects an abnormally high volume of configuration changes (15 or more within a 10-minute window) occurring on network appliances such as firewalls or security gateways, originated by a single source user or IP address. This behavior may indicate unauthorized administrative activity, mass policy updates, or malicious tampering.
Detects the addition of suspicious, attacker-controlled URLs (such as webhook.site, ngrok.io, or pipedream.net) as valid redirect URIs in an OAuth application's configuration. This technique can be used by an adversary to intercept OAuth authorization codes or tokens by luring users through a malicious authorization link, enabling persistent unauthorized access without further consent.
This rule detects when new credentials (secrets or certificates) are added to an Azure Active Directory Service Principal. It flags this activity as suspicious if it occurs outside of typical business hours or if it exhibits signs of 'rapid re-keying', where multiple credentials are added to the same service principal within a 24-hour window. This behavior is a common indicator of persistence maintenance or credential rotation to support unauthorized access.
Page 3 of 40