Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,153 detections
Filters
Last updated
All Time
Detection languages
541
211
97
75
69
Contributors
363
214
171
65
55
Categories
384
376
337
250
233
Platforms
1,097
44
8
7
3
Products / Services
448
379
89
86
36
MITRE Techniques
18,123
15,639
12,581
8,394
6,215
1,153
CVEs
6
5
2
1
1
IDS Classtypes
3
2
2
IDS Protocols
4
1
Detects suspicious cross-process access or injection activity targeting Internet Explorer-related processes (iexplore.exe, ieinstal.exe, ielowutil.exe). The rule monitors for high-privileged access rights, non-standard source process paths, and the presence of memory-related API calls often associated with process hollowing or injection (e.g., ZwCreateSection, VirtualAlloc, SetThreadContext) in combination with repeated access attempts.
This rule monitors for Cobalt Strike Beacon named pipe C2 patterns by analyzing Sysmon event IDs 17 (Pipe Created) and 18 (Pipe Connected). It specifically flags known default Cobalt Strike pipe naming conventions (e.g., 'MSSE-', 'postex_ssh_', 'msagent_') and common randomized short alphanumeric pipe names while excluding legitimate Windows system pipes.
Detects instances where a process attempts to access the memory or control flow of a web browser (Chrome, Edge, Brave) using suspicious access rights (0x820), which correspond to PROCESS_VM_WRITE (allowing memory modification) and PROCESS_SUSPEND_RESUME (allowing thread suspension). This behavior is often indicative of process injection or hollowed execution attempts against browser processes.
Detects the initialization phase of the Operation DragonReturn DcRAT loader, specifically identifying the creation of a unique named global mutex (Global\ShitSetupOn26126k) which acts as a single-instance enforcement mechanism and is often associated with sandbox evasion techniques.
Detects cross-process injection attempts into the Windows system process 'svchost.exe' where the source process is either unsigned or lacks a verified Microsoft publisher. This pattern is commonly observed in malware payloads, such as DcRAT, attempting to hide malicious activity within a legitimate system process.
Detects the deployment and activity of the Remcos Remote Access Trojan (RAT) as part of the BQTLock (BaqiyatLock) attack chain. This rule monitors for process injection into critical Windows processes, malicious Run key persistence entries, non-standard outbound TLS communication from non-browser processes, and the creation of potential keylogging artifacts in the AppData directory.
This rule detects common activity patterns associated with the Havoc C2 framework and general malicious PowerShell usage. It monitors for three primary indicators: VBScript stagers spawning PowerShell processes (e.g., in a Havoc-like delivery chain), PowerShell processes initiating DNS queries to DuckDNS infrastructure, and PowerShell processes performing cross-process memory injection indicative of shellcode execution (e.g., Demon in-memory agent).
Detects activity associated with the SPECTRALVIPER malware loader, specifically identifying known malicious file hashes (system.config.xml, NotificationConfig.json) and cross-process memory injection patterns (RemoteMemoryAllocation, RemoteMemoryWrite, CreateRemoteThreadApiCall) initiated by specific suspect processes.
Detects reflective process injection activity where the IntelAudioService.exe process opens a handle to the OneDrive.Sync.Service.exe process, potentially using XGU::Manager::ProcessReflector class methods to inject shellcode.
Detects HTTP response bodies containing common PE/Shellcode magic headers (MZ, PE, or NOP sleds) which are associated with the delivery of payloads designed to exploit or utilize the Windows KernelCallbackTable for process injection. The detection focuses on suspicious content delivered over HTTP connections.
Detects the AdaptixC2 agent (AdgNsy.exe) performing process injection into the Windows Address Book (WAB.EXE). The rule specifically monitors for the creation of AdgNsy.exe, its execution under the WMI Provider Host (WmiPrvSE.exe), and subsequent cross-process access to WAB.EXE, which is a known behavior of the AdaptixC2 framework utilized during Akira ransomware intrusions.
Detects suspicious process injection attempts where common scripting/proxy execution tools (wscript.exe, cscript.exe, mshta.exe, powershell.exe, msiexec.exe) are used to inject into other processes (RegAsm.exe, InstallUtil.exe, vbc.exe, svchost.exe, explorer.exe) using CreateRemoteThread. The detection is further refined by looking for command-line indicators often associated with malicious activity, such as downloading content or executing scripts.
Detects instances where the WinRAR archive utility spawns a PowerShell process with command-line arguments typically associated with obfuscated or fileless execution. The detection specifically flags PowerShell commands that employ techniques such as base64 encoding, execution policy bypass, hidden window styles, or commands containing indicators of in-memory process manipulation such as NtAllocateVirtualMemory or memory allocation constants.
EXPERIMENTAL: This rule detects RoguePlanet indicators when the Microsoft Defender Antimalware Service Executable (MsMpEng.exe) creates or modifies an executable (.exe, .dll, .sys) file within the System32 or SysWOW64 directories. This behavior can indicate a potential local privilege escalation (LPE) or other malicious activity, especially if Defender is being abused for remediation in a SYSTEM context. The rule specifically highlights 'wermgr.exe' as a high-interest target, but is generalized to catch any PE file.
Detects DcRAT (Operation DragonReturn) malware beaconing to a known C2 IP address (223.26.63.40) following the execution of suspicious sideloaded binaries, specifically 'Mixed Reality.exe' or 'COU_ITR-1_to_4_AY2026-27.exe', which then inject malicious code into svchost.exe.
Detects when common Microsoft Office applications (Word, Excel, PowerPoint, Outlook) load a DLL file from suspicious or user-writable locations such as AppData, Downloads, Documents, or Temp folders. The rule specifically alerts on DLLs that are unsigned or where the loading process itself is unsigned, which is a common indicator of side-loading or malicious library injection.
Detects multiple suspicious activities associated with the AnyDesk remote access application, including the creation of specific named pipes, execution of processes with sensitive command-line arguments (e.g., comsvcs, impersonation privileges), non-standard DLL loads by the application, and the spawning of system-level processes. These behaviors are indicative of abuse of remote access software for post-exploitation activities or privilege escalation.
Detects the execution of MAVInject.exe (Microsoft Application Virtualization Injector) with the /INJECTRUNNING parameter, a technique used to inject malicious code into a running process to evade detection.
Infostealer distributed via fake open-source tool impersonation sites
Phishing-as-a-Service kit abusing OAuth device code flow against Microsoft 365
Fraud framework powering scam domains for crypto and gambling schemes
Page 24 of 55