Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
18,128 detections
Filters
Last updated
All Time
Detection languages
6,549
6,208
1,473
1,005
661
Contributors
3,555
2,652
2,153
2,087
1,515
Categories
7,572
3,901
1,465
1,393
1,330
Platforms
13,733
1,392
1,319
824
738
Products / Services
3,828
3,137
2,448
912
908
MITRE Techniques
18,128
15,651
12,599
8,410
6,235
CVEs
19
18
17
16
14
IDS Classtypes
305
51
24
23
22
IDS Protocols
308
136
13
10
10
This rule detects anomalous access to cloud services, specifically Salesforce, Drift, Salesloft, HubSpot, and Marketo, that may indicate OAuth token abuse. It identifies scenarios where a high volume of actions (8 or more) occur within a short timeframe (3 hours or less) from a single IP address but with multiple unique user agents (2 or more). The detection is further refined by looking for actions related to 'Add delegated permission grant' or 'Create service principal', which are indicative of an attacker attempting to establish persistence or escalate privileges after gaining access through a compromised OAuth token.
Detects common anti-forensic staging patterns used by IoT botnets such as TerraBot, r00ts3c, and RondoDox. The rule identifies processes that stage payloads in the /tmp directory, set world-writable execution permissions (chmod 777), and immediately delete artifacts following execution to hinder forensic analysis.
This rule detects attempts to impair defensive mechanisms by creating or modifying Quality of Service (QoS) policies that specifically target Endpoint Detection and Response (EDR) or Antivirus (AV) processes. Adversaries may use QoS policies to throttle network traffic for security products, hindering their ability to communicate with central management or cloud services, thus impairing their effectiveness. The rule looks for command-line arguments used to create or set QoS policies (`New-NetQosPolicy`, `Set-NetQosPolicy`) that include throttling parameters (`ThrottleRateActionBitsPerSecond`, `AppPathNameMatchCondition`) and target common EDR/AV process names. It also detects direct registry modifications to QoS policies (`\SOFTWARE\Policies\Microsoft\Windows\QOS`) that include EDR/AV process names.
Detects the use of esxcli or vim-cmd commands on VMware ESXi hosts to power off virtual machines, remove software components (VIBs), or enable maintenance mode. These actions are commonly performed by the Akira ransomware group to disrupt operations and facilitate encryption.
Detects the loading of known vulnerable or unsigned drivers, specifically targeting Huawei-signed driver variants, which have been leveraged by DragonForce for Bring Your Own Vulnerable Driver (BYOVD) defense evasion. The rule identifies driver loads where the file name matches known vulnerable Huawei drivers or where the driver is loaded by an unusual process and not from a standard system directory, and is also untrusted.
This rule detects the loading of specific DLLs associated with Remcos RAT (Optimax.dll, System Optimizer, SystemOptimizer) from temporary directories. This behavior is indicative of in-memory execution patterns often used by malware to evade detection.
This rule detects suspicious executable or DLL file creations in common temporary or user-writable directories shortly after a Microsoft Defender related process (MsMpEng, Nissrv, Werfault, Services) terminates unexpectedly. The rule focuses on the causality chain to link the Defender crash with the subsequent file placement, suggesting an attempt to bypass or disable security software before deploying malicious payloads.
Detects the use of command-line arguments associated with Windows token impersonation techniques (e.g., 'SeImpersonate', 'ImpersonateLoggedOnUser') initiated by common command-line interpreters (powershell.exe, cmd.exe, wmic.exe). This activity is often indicative of an adversary attempting to elevate privileges by stealing or manipulating access tokens.
This rule monitors registry modifications for keys or values related to common virtualization platforms (VirtualBox, VMware, QEMU, Xen, Hyper-V). Such registry queries or modifications are often indicative of malware performing sandbox evasion to determine if the execution environment is a virtual machine or a sandbox used for analysis, allowing the malware to terminate or alter its behavior to avoid detection.
Detects the execution of executable files with suspicious command line arguments related to AI or Machine Learning frameworks (e.g., tensorflow, pytorch, neural) where the file size is relatively small (between 50KB and 500KB), which may indicate potentially malicious payloads masquerading as legitimate AI utilities.
This rule detects the execution of the squid.exe process where the command line does not explicitly specify a configuration file using the '-f' or '-n' flags, nor does it reference 'squid.conf'. Running Squid without a defined configuration file may lead to the application loading default, insecure, or unintended settings, which could be exploited as part of a vulnerability chain or misconfiguration.
Detects an SSPR (Self-Service Password Reset) event immediately followed by the registration of a new Microsoft Authenticator method for the same user. This sequence of events can indicate an attacker has compromised an identity, performed a password reset, and is attempting to establish a new MFA method for persistent access.
Detects the instantiation of the Shell.Explorer.1 COM object via OLE in RTF documents, which is indicative of an attempt to exploit CVE-2026-21509 (PRISMEX). This vulnerability allows for potential arbitrary code execution.
Detects the MYRA RAT using memfd_create and fexecve to execute a Rust payload entirely in memory without writing an executable to disk, as observed in the Operation Friday Harvest npm supply chain campaign. The rule looks for process events where the action type indicates process creation, memory allocation, or process access, or where the FolderPath starts with '/proc/' and contains '/fd/'. It specifically targets initiating processes named 'node' or 'npm' or those with command lines associated with npm post-installation scripts. Further filtering is applied to FolderPaths matching the '/proc/\d+/fd/\d+' regex or ProcessCommandLines containing 'memfd_create' or 'fexecve', or FolderPaths containing 'memfd'.
Detects the creation or modification of registry values used by the HIDDENDRIVER rootkit to configure its hiding and protection capabilities. This rootkit, a modified version of the open-source "Hidden" project, uses these registry entries to define which files, processes, and registry keys to conceal. This activity is associated with the REF3927 threat actor and the TOLLBOOTH malware campaign.
Detects behaviors associated with AMOS and Odyssey Stealer, including using osascript to perform anti-analysis checks via system_profiler and terminating OneDrive services to impair defenses.
Detects the creation or modification of registry values used by the HIDDENDRIVER rootkit to configure its hiding and protection capabilities. This rootkit, a modified version of the open-source "Hidden" project, uses these registry entries to define which files, processes, and registry keys to conceal. This activity is associated with the REF3927 threat actor and the TOLLBOOTH malware campaign.
Detects DbgView.exe, DbgView64.exe, or VirtualBox.exe processes making unexpected outbound network connections to public IP addresses. This behavior is associated with Backdoor.Turn, which is known to inject into these trusted processes via DLL sideloading to communicate with C2 servers, potentially routing traffic through Microsoft Teams TURN relays.
Detects the creation of a child process by a known analysis, debugging, or reverse engineering tool. Malware like Stealit checks for these parent processes to determine if it's being analyzed. Spawning unexpected child processes from these tools can indicate malicious activity, such as the malware continuing its execution path after a failed anti-analysis check or a user interacting with a malicious file within a sandbox. The rule is filtered to exclude common legitimate child processes spawned by these tools.
Detects the creation of a child process by a known analysis, debugging, or reverse engineering tool. Malware like Stealit checks for these parent processes to determine if it's being analyzed. Spawning unexpected child processes from these tools can indicate malicious activity, such as the malware continuing its execution path after a failed anti-analysis check or a user interacting with a malicious file within a sandbox. The rule is filtered to exclude common legitimate child processes spawned by these tools.
Detects the execution of executables with specific names ("WinHealth.exe", "WindowsSecurity.exe") known to be dropped by the SonicCrypt crypter. The execution is often initiated by the Task Scheduler, resulting in a parent process of 'taskhostw.exe'.
Page 400 of 864