Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,733 detections
Filters
Last updated
All Time
Detection languages
909
387
164
127
68
Contributors
479
417
201
177
61
Categories
20,440
11,814
5,871
5,145
4,864
1,733
Platforms
822
436
329
274
206
Products / Services
498
378
118
96
79
MITRE Techniques
433
327
225
196
196
CVEs
12
12
10
10
6
IDS Classtypes
1
1
IDS Protocols
2
Detects scenarios where an attacker grants new privileges on a database level
Detects scenarios where an attacker deactivates SQL Server database auditing capacities. SQL auditing requires previous configuration on each SQL instance.
Detects scenarios where an attacker deactivates SQL Server auditing capacities. SQL auditing requires previous configuration on each SQL instance.
Detects scenarios where an attacker adds his account to a database role (db_securityadmin, db_datawriter, db_securityadmin, ...)
Detects a user self-assigning access to a large number of applications in Okta within a short time frame. This behavior is associated with the threat group UNC3944 (Scattered Spider) for privilege escalation and lateral movement across SaaS applications.
This Logic detects the presence of the malicious Chrome extension FreeVPN.One by identifying its unique extension ID across installed browsers. The logic further correlates this presence with network communications initiated by the extension to suspicious or untrusted domains. By combining extension enumeration with traffic analysis, the detection ensures high fidelity with minimal false positives. This layered approach strengthens visibility into malicious browser add-ons masquerading as VPN tools.
This rule detects the presence of potentially malicious or unwanted browser extensions and outbound network connections to known adware or tracking domains. It correlates extension presence with network behavior to identify threats.
This rule detects the installation of potentially unwanted or suspicious applications, specifically looking for a sponsored version of FileZilla. It then correlates this installation with outbound network activity to public IP addresses, which could indicate telemetry, beaconing, data exfiltration, or command and control (C2) behavior.
The following analytic detects attempts to install unsigned MSIX/AppX packages using the -AllowUnsigned parameter. This detection leverages Windows event logs from the AppXDeployment-Server, specifically focusing on EventID 603 which indicates the start of a deployment operation with specific deployment flags. The flag value 8388608 corresponds to the -AllowUnsigned option in PowerShell's Add-AppxPackage cmdlet. This activity is significant as adversaries have been observed leveraging unsigned MSIX packages to deliver malware, bypassing signature verification that would normally protect users from malicious packages. If confirmed malicious, this could allow attackers to execute arbitrary code, establish persistence, or deliver malware while evading traditional detection mechanisms.
This analytic detects the installation of MSIX/AppX packages with full trust privileges. This detection leverages Windows event logs from the AppXDeployment-Server, specifically focusing on EventCode 400 which indicates a package deployment operation. Full trust packages are significant as they run with elevated privileges outside the normal AppX container restrictions, allowing them to access system resources that regular AppX packages cannot. Adversaries have been observed leveraging full trust MSIX packages to deliver malware, as documented in recent threat intelligence reports. If confirmed malicious, these packages could allow attackers to execute arbitrary code with elevated privileges, establish persistence, or deliver malware while evading traditional detection mechanisms.
This detection identifies the installation of developer-signed MSIX packages that lack Microsoft Store signatures. All malicious MSIX packages observed in recent threat campaigns (including those from FIN7, Zloader/Storm-0569, and FakeBat/Storm-1113) were developer-signed rather than Microsoft Store signed. Microsoft Store apps have specific publisher IDs containing '8wekyb3d8bbwe' or 'cw5n1h2txyewy', while developer-signed packages lack these identifiers. This detection focuses on EventID 855 from the Microsoft-Windows-AppXDeployment-Server/Operational logs, which indicates a completed package installation.
This analytic detects successful MSIX/AppX package installations on Windows systems by monitoring EventID 854 in the Microsoft-Windows-AppXDeployment-Server/Operational log. This event is generated when an MSIX/AppX package has been successfully installed on a system. While most package installations are legitimate, monitoring these events can help identify unauthorized or suspicious package installations, especially when correlated with other events such as unsigned package installations (EventID 603 with Flags=8388608) or full trust package installations (EventID 400 with HasFullTrust=true).
This hunting query detects user interactions with MSIX packages by monitoring EventCode 171 in the Microsoft-Windows-AppXPackaging/Operational logs. These events are generated when a user clicks on or attempts to interact with an MSIX package, even if the package is not fully installed. This information can be valuable for security teams to identify what MSIX packages users are attempting to open in their environment, which may help detect malicious MSIX packages before they're fully installed. Monitoring these interactions can provide early warning of potential MSIX package abuse, which has been leveraged by threat actors such as FIN7, Zloader (Storm-0569), and FakeBat (Storm-1113).
Detects attempts to perform prompt injection attacks against Large Language Models (LLMs) and AI bots. These attacks aim to subvert the model's original instructions, bypass safety mechanisms, or extract sensitive information.
This rule detects an impersonation event within Microsoft Teams, specifically when the 'TeamsImpersonationDetected' action type is logged. It extracts the User ID and the Sender's User Principal Name (UPN) associated with the detected impersonation.
This rule identifies devices that have software installed which has reached its End-Of-Support (EOS) status. This is crucial for identifying potential security vulnerabilities due to unpatched software and for maintaining compliance with organizational security policies. The rule summarizes the count of such software instances per device, along with the software name and its End-Of-Support date.
This analytic leverages alerts from Splunk AppDynamics SecureApp to identify and monitor exploit attempts targeting business applications. It specifically looks for vulnerabilities in web applications, including injection attacks (SQL, API abuse), deserialization vulnerabilities, remote code execution attempts, LOG4J, and zero-day attacks. These attacks typically aim to gain unauthorized access, exfiltrate sensitive data, or disrupt application functionality. Splunk AppDynamics SecureApp provides real-time detection by analyzing application-layer events and correlating attack behavior with known vulnerability signatures, categorizing attack severity, providing visibility into attacker tactics, and generating risk-based scoring and contextual alerts.
This rule identifies software installed on devices that has reached its End of Support (EOS) version. It provides details such as the device name, software name, vendor, version, EOS status, and EOS date, which is crucial for identifying potential security risks due to unpatched or unsupported software.
This rule identifies software installed on devices that is approaching its end-of-support (EOS) date. It queries the DeviceTvmSoftwareInventory table to find software with an 'Upcoming EOS Version' status, providing details such as device name, software name, vendor, version, EOS status, and EOS date. This can help in proactive patch management and identifying potential security risks associated with outdated software.
Indicates suspicious rules such as an inbox rule that forwards a copy of all emails to an external address
This detection identifies when a one-on-one Microsoft Teams chat is initiated with an external domain that has not been observed in previous Teams chats (past 90 days) or recent email communications (past 15 days). This may indicate a new, potentially unsolicited or suspicious contact attempting to engage with an internal user, which could represent a social engineering attempt or early-stage reconnaissance.
Page 72 of 83