Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
20,482 detections
Filters
Last updated
All Time
Detection languages
8,114
7,158
1,351
1,168
1,063
Contributors
4,878
2,724
2,695
2,646
2,356
Categories
20,482
11,866
5,884
5,164
4,867
Platforms
18,085
3,046
1,506
124
104
Products / Services
5,725
4,293
3,715
1,174
958
MITRE Techniques
9,480
7,573
3,458
2,848
2,702
CVEs
36
28
25
21
20
IDS Classtypes
1
Detects the use of the SharpGPOAbuse tool, typically used for modifying Group Policy Objects for persistence or privilege escalation, in conjunction with rapid Active Directory group enumeration (using net.exe or net1.exe) on the same host within a 5-minute window. This behavior is highly indicative of an adversary performing post-exploitation reconnaissance and privilege escalation activities.
Detects unauthorized inbound WebSocket connections from external (non-private) IP addresses to the AutoGen Studio MCP API endpoint at /api/mcp/ws/ on port 8081, followed by the execution of interpreter processes (autogenstudio, uvicorn, or python) on the same host, indicating a potential exploitation of an origin validation bypass for remote code execution.
This rule detects instances where the Windows Update service (wusvc.exe) loads a module or file while simultaneously initiating a process. This behavior is often associated with process injection or the misuse of legitimate system binaries (LOLBins) to facilitate malicious activities by proxying execution through trusted services.
This rule monitors for the execution of 'wusvc.exe' (a potential disguise for malicious binaries) when performed by users or on hostnames containing keywords associated with security research, sandboxing, or testing such as 'sandbox', 'malware', 'virus', or 'sample'.
Detects the execution of VBScript files containing 'ragecore' in the command line, specifically when launched via wscript.exe or cscript.exe from common staging or persistence directories such as Windows Temp, Users Public, or Startup folders.
Detects file write or delete operations involving cmd.exe by processes potentially associated with the TrymeLocker ransomware, specifically looking for indicators of ownership or permission modification (takeown, icacls) and suspicious file paths.
Detects the use of native Windows utilities (icacls.exe, cacls.exe, takeown.exe) to modify file or directory permissions or ownership in a manner consistent with adversarial activity. This includes granting full permissions to everyone/administrators, or changing ownership of critical system directories (e.g., windows, system32) to standard user accounts. The rule includes logic to filter out known legitimate administrative processes and flags potential suspicious parent processes.
This rule detects the execution of processes named win_guard_*.exe or those residing in directories often used for malicious staging (Users\Public, Windows\Temp, ProgramData). Additionally, it detects when these identified suspicious processes load user32.dll, a common indicator of initiating a low-level keyboard hook (WH_KEYBOARD_LL) often used by keyloggers to capture user input.
This rule detects potential command and control (C2) activity associated with the SNOWLIGHT framework by identifying outbound WebSocket connections over HTTP to a specific known C2 IP (38.60.253.35) and port (8084). It correlates network flow logs (CommonSecurityLog) containing specific HTTP URL parameters (a=w64, t=ws_, p=8084) with EDR process network events (DeviceNetworkEvents) where the Windows Update Service process (wusvc.exe) is the initiator, indicating potential process masquerading or injection.
Detects an unknown/network-active parent process spawning shells used by ArachneC2 interactive shell or exec functionality.
Detects suspicious execution patterns involving Windows file management utilities (takeown.exe, icacls.exe) or the deletion of command-line interpreter binaries through cmd.exe. These patterns are often associated with unauthorized modifications of file permissions or attempts to clear forensic evidence during an intrusion.
Detects multiple execution attempts of common Windows built-in utilities (bcdedit, vssadmin, wbadmin, wmic) used to disable recovery features or delete volume shadow copies within a 5-minute window. This behavior is highly characteristic of ransomware or destructive malware aiming to prevent system recovery.
Detects evidence of RustDesk session permission bypass exploitation. The rule monitors for the presence of known Proof-of-Concept (PoC) file artifacts, suspicious child processes spawned by RustDesk, use of elevated tokens by RustDesk-initiated processes, compilation of PoC tools using cargo, and outbound network connections from RustDesk to known relay ports.
Detects multiple concurrent attempts to inhibit Windows system recovery, a common behavior observed in the TrymeLocker ransomware family. The rule triggers when at least two distinct malicious activities occur on the same host within a 120-second window, including the disabling of the Windows Recovery Environment (WinRE) via bcdedit, deletion of shadow copies via vssadmin or wmic, disabling of the Windows backup service (wbengine) via sc, or disabling of System Restore via registry modification.
This rule detects potentially unauthorized installation of extensions in Visual Studio Code (VSIX) or plugins in JetBrains IDEs (IntelliJ, PyCharm, WebStorm). It flags installations that do not originate from verified official marketplaces, which may indicate the manual installation of malicious or supply-chain compromised development environment extensions.
This rule detects the use of the 'powercfg.exe' utility by non-system accounts to modify power settings, such as disabling hibernation, modifying standby/sleep timeouts, or changing monitor/disk timeout configurations. Such actions can be indicative of attempts to maintain system availability, prevent the system from entering a low-power state that might terminate malicious processes, or ensure persistent execution of unauthorized activities.
Detects the spawning of shell processes (sh, bash, dash, zsh, ksh) as children of ffmpeg, ffprobe, jellyfin-ffmpeg, or emby-ffmpeg. This behavior is indicative of a successful AVBuffer.free pointer hijack vulnerability (CVE-2026-8461, also known as PixelSmash) being exploited to achieve arbitrary command execution. The rule specifically looks for these parent processes spawning shell children, or if the child process command line contains common shell paths or network-related commands, and filters for specific user accounts often associated with media servers or generic system accounts.
Detects instances where rundll32.exe is used to execute a file with a suspicious non-DLL extension (such as .tmp, .dat, .bin, .jpg, etc.) by specifying an exported function via an ordinal number. Adversaries often use this technique to proxy malicious code execution while attempting to evade security tools that only look for standard .dll extensions.
Detects the use of PowerShell to perform DNS lookups for TXT records followed by immediate command execution, such as using 'Invoke-Expression' or 'Start-Process'. This pattern is frequently indicative of fileless command-and-control (C2) communication where stage-payload commands or scripts are retrieved via DNS TXT records to evade detection.
Detects the execution of Python interpreters spawned by code editors or runtime environments (e.g., VS Code, Cursor, Node.js) with specific command-line arguments involving 'init'. This behavior may indicate an adversary attempting to leverage development tools to execute custom scripts, potentially for automated deployment, persistence, or malicious payload initialization.
Detects instances of rundll32.exe being executed with command-line arguments that suggest malicious activity, specifically loading modules from remote UNC paths, using the javascript: protocol, or explicitly invoking ShellExec_RunDLL via shell32.dll. This rule filters out legitimate signed Microsoft binaries to focus on potentially unauthorized or malicious proxies.
Page 457 of 976