Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
9,759 detections
Filters
Last updated
All Time
Detection languages
3,724
2,036
1,335
1,055
969
Contributors
2,963
2,050
1,372
1,226
485
Categories
5,718
3,075
1,138
900
659
Platforms
9,646
658
432
53
30
Products / Services
10,696
9,759
6,696
4,421
3,752
MITRE Techniques
3,828
3,376
1,980
1,617
1,042
CVEs
18
14
12
12
11
Detects the copying of the legitimate Windows curl.exe binary from System32 to a user-writable directory (such as AppData\Local\Temp) and renaming it. This is a common LOLBin (Living Off the Land Binary) technique used by adversaries to disguise the utility as a benign application to evade detection while facilitating the download of second-stage payloads.
Detects the execution of the Pester framework's Build.bat script, which is being abused as a LOLBAS (Living Off the Land Binary and Script) technique. This activity is typically characterized by being launched from an LNK file, often indicating a malicious dropper or secondary stage payload delivery.
Detects a Node.js server process (e.g. "next-server", "node.exe") spawning
cmd.exe with the "/d /s /c" flag combination, which is the pattern
produced by Node's child_process.execFileSync("cmd.exe", ["/d","/s","/c",
...]) — the exact command shape used by the CVE-2026-75604 exploit chain
to run attacker-supplied commands after a Server Action deserialization
bypass. Node applications almost never legitimately spawn cmd.exe.
cmd.exe with the "/d /s /c" flag combination, which is the pattern
produced by Node's child_process.execFileSync("cmd.exe", ["/d","/s","/c",
...]) — the exact command shape used by the CVE-2026-75604 exploit chain
to run attacker-supplied commands after a Server Action deserialization
bypass. Node applications almost never legitimately spawn cmd.exe.
Detects potential process injection attempts by unidentified executable files originating from suspicious directories (AppData or Temp) targeting critical system or service processes (e.g., explorer.exe, svchost.exe). The rule monitors CreateRemoteThread events and excludes signed dotnet host processes, characteristic of Hannibal Stealer behavior.
Detects instances where a process other than the legitimate Firefox browser executable loads the NSS (Network Security Services) libraries nss3.dll or mozglue.dll from the Firefox installation directory. This behavior is indicative of unauthorized attempts to decrypt saved browser credentials by interacting with Firefox-specific security modules, a technique commonly used by credential-stealing malware.
Detects the creation of a shortcut (LNK) file within the Windows Startup directory that points to a msedge.exe binary located in a non-standard, masqueraded location within the user's Local AppData path. This behavior is associated with the LucidPawn/LucidRook persistence mechanism.
Detects the execution of the Chisel C2 tool by matching its unique command-line grammar, specifically the 'client' subcommand combined with reverse remote ('R:') configurations. This detection strategy is resilient against binary renaming and infrastructure rotation by focusing on the tool's required argument structure rather than static IOCs like filenames or C2 addresses. It is specifically useful for identifying unauthorized remote tunnels used for persistence and data exfiltration.
Detects MayaBot's scheduled-task persistence chain by correlating two signals on the same device within a 60-minute window: (1) schtasks.exe creating a daily task referencing a hidden batch script (update.bat/backup.bat/firewall-update.bat) under AppData\Local, spawned from wscript.exe/cscript.exe/cmd.exe, and (2) cmd.exe launching a hidden-window PowerShell process referencing the same batch scripts under AppData\Local. Requiring both signals together, anchored to the AppData\Local path, sharply reduces false positives compared to alerting on either behavior alone.
Detects MayaBot's scheduled-task persistence chain by correlating two signals on the same device within a 60-minute window: (1) schtasks.exe creating a daily task referencing a hidden batch script (update.bat/backup.bat/firewall-update.bat) under AppData\Local, spawned from wscript.exe/cscript.exe/cmd.exe, and (2) cmd.exe launching a hidden-window PowerShell process referencing the same batch scripts under AppData\Local. Requiring both signals together, anchored to the AppData\Local path, sharply reduces false positives compared to alerting on either behavior alone.
Detects instances where PowerShell scripts are executed via the Windows Task Scheduler from within potential staging directories (ProgramData with randomized folder names). The rule identifies execution that uses common PowerShell obfuscation flags (e.g., -EncodedCommand) and requires secondary hardening bypass flags (e.g., -WindowStyle hidden), followed by correlated network activity to a specific suspicious C2 domain.
This rule detects potential PowerShell script execution without spawning the standard powershell.exe process. It monitors for the loading of clr.dll and System.Management.Automation.dll into any process other than powershell.exe, which is a technique used by some malware families like SloppyRAT to execute PowerShell code directly within a compromised process memory space.
Detects the indexed-btree npm malware loader embedded in BTree.prototype.set that spawns a detached hidden Node.js child process and references sharedLoad.min.js
Detects the spawning of command-line interfaces such as cmd.exe, powershell.exe, or pwsh.exe as child processes of the SharePoint web application worker process (w3wp.exe). This behavior is highly irregular for a web server process and is indicative of potential exploitation, such as remote code execution (RCE) via web application vulnerabilities.
Detects the deletion of .zip files within the Public directory by VBScript or JScript interpreters (wscript.exe or cscript.exe). This pattern is often indicative of a malicious dropper or downloader that extracts a payload from a staged archive and subsequently removes the evidence to hinder analysis and forensics.
Detects the execution of Windows Script Host (wscript.exe/cscript.exe) loading WMI-related modules (wbemprox.dll, wbemcomn.dll, etc.). This pattern is often indicative of scripts attempting to query WMI for system information, environment configuration, or locale data, which is a common reconnaissance technique used by malware to identify the victim environment.
This rule detects the execution of script files (such as .vbs, .vbe, .js, .jse) via wscript.exe when the file name uses a double-extension technique (e.g., .pdf.vbs) to masquerade as a benign PDF document. These files are typically found in temporary or user-download directories, indicating potential malicious activity initiated by downloaded attachments.
Detects the launch of client32.exe (a core component of NetSupport Manager) as a child process of PowerShell, indicating potential malicious deployment of the software as a RAT rather than legitimate administrative activity.
Detects a suspicious sequence of user activity where a device visits a social media platform (Twitter/X) followed shortly by navigation to a Google Doc. This pattern is often consistent with phishing lures where victims are enticed via direct messages or social engineering to open a malicious document hosted on Google Docs.
Detects the presence and execution of components related to the 'GhostCode' activity, specifically the installation of a rogue self-signed root certificate authority (CA) masquerading as Google Trust Services and the subsequent configuration of a local TLS-intercepting proxy. The detection monitors for the creation of specific archive files, modification of the system certificate store, addition of local firewall rules, and modification of the local hosts file.
Detects a suspicious sequence where a Python process is spawned by a 'marimo' notebook process, followed immediately by an outbound SSH connection attempt from the same device. This pattern is indicative of automated or scripted lateral movement where a notebook environment is abused as a jump-point for command execution on remote systems.
Detects the specific sequence of administrative utilities used by Gentlemen ransomware to seize control of files before encryption. This involves using 'takeown.exe' to claim ownership, 'icacls.exe' to grant full access to the Everyone group, and 'attrib.exe' to remove read-only file attributes, a common precursor to mass file encryption.
Page 68 of 465