Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
20,483 detections
Filters
Last updated
All Time
Detection languages
8,114
7,158
1,351
1,169
1,063
Contributors
4,878
2,724
2,695
2,646
2,356
Categories
20,483
11,867
5,886
5,169
4,867
Platforms
18,085
3,047
1,506
124
104
Products / Services
5,726
4,293
3,715
1,174
958
MITRE Techniques
9,480
7,574
3,458
2,848
2,703
CVEs
36
28
25
21
20
IDS Classtypes
1
This rule detects a sequence of suspicious activities: the creation or modification of an executable file (.exe or .dll) within a directory path containing update-related keywords, followed by the execution of a file in that same location, and culminating in an outbound network connection from that process. This pattern is characteristic of a software supply chain compromise or an adversary deploying a malicious payload via a fake or compromised update mechanism.
Detects the use of PowerShell, PWSH, or CMD to modify file system timestamps, a technique commonly referred to as timestomping. This is often used by adversaries to hide malicious file activity or blend in with legitimate system files by manipulating file creation, modification, or access time metadata.
This rule detects potential persistence mechanisms associated with Microsoft Outlook, specifically targeting registry modifications related to Outlook security settings or suspicious child processes (cmd.exe or powershell.exe) spawned by outlook.exe.
Detects instances where a non-SYSTEM parent process spawns a child process running as SYSTEM, characteristic of 'Potato' family exploits (e.g., JuicyPotato, PrintSpoofer) leveraging SeImpersonatePrivilege to elevate privileges. The rule specifically filters for known legitimate SYSTEM process spawners while highlighting suspicious parent-child process relationships.
Detects unauthorized attempts to enumerate stored Windows credentials using native command-line utilities such as vaultcmd.exe and cmdkey.exe, or by invoking CredEnumerate/Get-StoredCredential functions via PowerShell scripts. These methods are commonly used by attackers to gain access to cached passwords, tokens, or network credentials stored in the Windows Credential Manager.
This query detects the creation of local user accounts on Linux systems using the 'useradd' or 'adduser' commands. Monitoring for local account creation is crucial as adversaries may create accounts for persistence, privilege escalation, or to facilitate lateral movement.
Detects the execution of PowerShell with a Base64-encoded command line originating from Windows Explorer, a behavior consistent with the Potemkin Loader and ClickFix attack campaigns used to bypass string-based detection.
Detects the Potemkin Loader using the native Windows curl.exe utility to download a ZIP archive from a remote source, followed by the creation of a local .zip file. This behavior is used to bypass security controls by avoiding PowerShell-based download mechanisms.
Detects persistence attempts by the Amadey infostealer. The rule monitors for unauthorized creation of .job files directly in the Windows Tasks directory or the execution of schtasks.exe from suspicious locations such as TEMP or APPDATA folders.
Detects the execution of PowerShell with the 'Expand-Archive' cmdlet originating from a suspicious or non-standard parent process. This pattern is commonly used by malware loaders, such as Amadey, to unpack second-stage payloads like StealC from downloaded ZIP archives hidden in temporary application folders.
Detects a multi-stage attack pattern involving a 'spam bomb' (high volume of inbound emails) followed by an external Microsoft Teams message and the subsequent execution of Windows Quick Assist (quickassist.exe or msra.exe) on the target endpoint. This behavior is highly indicative of a tech support scam or social engineering attack where an adversary floods a user's inbox to create urgency, contacts them via Teams, and convinces them to initiate a remote support session.
Hunting rule loosely based off of Atomic Red Team T1518.001 test "Get Windows Defender exclusion settings using WMIC". No public in-the-wild intel reference available; correlate before alerting.
Detects suspicious process activity originating from common build and package management tools (node, npm, bun, npx) that targets sensitive cloud and container credential locations, such as AWS IMDS, AWS configuration files, Kubernetes service account tokens, and HashiCorp Vault secrets. This rule is designed to identify malicious post-installation scripts or build processes attempting to exfiltrate credentials from developer or CI/CD environments.
Detects WMI-based process execution (T1047) where a script interpreter (wscript.exe/cscript.exe) executes a JavaScript or script file, followed within 30 seconds by WmiPrvSE.exe spawning PowerShell. This reconstructs the broken process chain created when scripts use WbemScripting.SWbemLocator and Win32_Process.Create() to proxy execution through WMI.
Why suspicious: Legitimate use of WMI to spawn interactive shells is rare. This technique is specifically used to sever the parent-child process relationship between the initial script host and PowerShell, evading parent-process-based EDR detections and allowlist rules.
Observed in: Remcos RAT campaigns using VHDX-packaged JavaScript loaders, but the technique is not exclusive to this malware family.
MITRE ATT&CK: T1047 - Windows Management Instrumentation
Why suspicious: Legitimate use of WMI to spawn interactive shells is rare. This technique is specifically used to sever the parent-child process relationship between the initial script host and PowerShell, evading parent-process-based EDR detections and allowlist rules.
Observed in: Remcos RAT campaigns using VHDX-packaged JavaScript loaders, but the technique is not exclusive to this malware family.
MITRE ATT&CK: T1047 - Windows Management Instrumentation
Detects anomalous child process execution (PowerShell, mshta, wscript, cscript, or cmd) spawned by common web browsers, which is consistent with the execution chain of CastleLoader malware delivered via EvilTokens PhaaS and ClickFix CAPTCHA lures. These lures typically trick users into copying and pasting commands into the Run dialog or directly into the shell.
Hunting rule for tar archive creation of user home trees, common staging before scp/rsync exfiltration over SSH. Prioritize /tmp archives and remote copy in the same session.
Hunting rule for tar archive creation of user home trees, common staging before scp/rsync exfiltration over SSH. Prioritize /tmp archives and remote copy in the same session.
Detects indicators associated with the deployment and execution of NetSupport RAT, including the launch of 'client32.exe', loading of the 'pcicl32.dll' module, outbound network connections, and the creation of the 'client.ini' configuration file. It also flags execution chains where 'client32.exe' is spawned by common browser or living-off-the-land processes.
Detects evidence of credential dumping activities, including the execution of known Mimikatz-related binaries or command lines, unauthorized memory dumping of the LSASS process using tools like procdump or rundll32 with comsvcs.dll, and the presence of known credential dumping artifacts on disk. It also monitors cross-process access to lsass.exe by potentially unauthorized or unsigned processes.
Detects potential timestomping activities via PowerShell APIs or touch.exe usage, alongside the creation or modification of known webshell filenames in web directories or files placed in common web root subdirectories by web server processes.
Detects the use of WinRAR (rar.exe/winrar.exe) to create archives with password protection or recursive flags in common staging directories (Temp, AppData, ProgramData, Users\Public, Share). This behavior is often associated with the staging and obfuscation of collected data prior to exfiltration.
Page 418 of 976