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
This rule detects persistence mechanisms by monitoring additions to Windows Registry Run keys or files created in the user's Startup directory. It specifically flags entries that target suspicious locations (e.g., Temp, AppData, Public) or attempt to execute encoded commands using PowerShell or CMD.
Detects the use of token manipulation APIs (DuplicateToken, DuplicateTokenEx, ImpersonateLoggedOnUser), the 'runas' command for credential switching, or the execution of known security token manipulation tools (e.g., incognito, Tokenvator, Invoke-TokenManipulation). This activity is commonly associated with privilege escalation and token theft.
Detects suspicious cross-process memory operations where a source process requests process creation/write permissions on a target process (e.g., browsers, explorer.exe, or svchost.exe), followed immediately by the execution of a remote thread injection technique such as CreateRemoteThread, QueueUserAPC, or NtCreateThreadEx within the same process pair.
Detects suspicious processes such as PowerShell, Python, or scripting hosts performing network communication with major AI/LLM provider APIs, followed immediately by child process execution or related activity. This pattern is indicative of LLM-assisted automation for malicious activities, potentially using AI to generate code, scripts, or orchestrate command and control actions.
Detects a multi-stage evasion sequence where a process first queries the system for virtualization or sandbox artifacts (e.g., VM tools, system information), followed by an intentional sleep or delay to bypass sandbox analysis, and concluding with subsequent process activity. This pattern is characteristic of malware attempting to detect and evade automated analysis environments.
This rule detects potential process injection or process hollowing attempts by correlating Sysmon Event ID 10 (ProcessAccess) events indicating process memory modification permissions (such as PROCESS_VM_WRITE and PROCESS_CREATE_THREAD) with subsequent Sysmon Event ID 8 (CreateRemoteThread) events targeting the same process. This behavior is indicative of an adversary injecting malicious code into a legitimate host process (e.g., svchost.exe, explorer.exe).
This rule detects cross-process operations where common user applications or scripting hosts initiate activity towards sensitive system processes (e.g., lsass.exe, services.exe). This pattern is a frequent indicator of process injection techniques used to achieve code execution in the context of high-privilege or critical system services, often for the purpose of credential theft or persistence.
Detects the execution of known Living-off-the-Land Binaries (LOLBAS) like mshta, rundll32, regsvr32, and others, when the command line includes indicators of external network retrieval (http/ftp), script execution (javascript/vbscript), or common obfuscation/decoding patterns (base64, encodedcommand). The rule excludes paths within System32 and SysWOW64 to focus on potentially suspicious locations or renamed binaries.
Detects a ClickFix-style activity where an adversary uses nslookup.exe or certutil.exe to perform DNS TXT record queries for payload staging or C2 rendezvous. The rule identifies suspicious parent processes like explorer.exe or scripting hosts, correlates this with a subsequent outbound network connection from the same host, and is intended to capture DNS-based beaconing or data staging.
Detects a suspected ransomware double-extortion sequence: the bulk creation of password-protected archives using compression utilities (e.g., 7z, RAR), followed by significant outbound network connections to common cloud storage or anonymization endpoints, and concluding with mass file modifications indicative of encryption activity.
Detects the installation of known AI-powered coding assistant extensions (e.g., Copilot, Tabnine, Cursor) on a device, followed by significant source-code repository cloning or archival activity by Git within two hours. This pattern is potentially indicative of intellectual property theft or sensitive source code exfiltration facilitated by a newly introduced AI tool.
Detects the abuse of the legitimate Windows system binaries regsvr32.exe and rundll32.exe to execute remote scripts or scriptlets, a technique often used to bypass application control and proxy execution of malicious code (similar to Squiblydoo).
Detects instances where a digitally signed executable loads a DLL from a non-standard, user-writable directory (such as Downloads, Desktop, Temp, or AppData). This behavior is characteristic of DLL side-loading, an evasion technique used to execute malicious code by placing a rogue DLL in a location where a legitimate binary might search for its dependencies, often used by threat actors to persist or maintain covert execution.
This rule detects artifacts and command-line patterns associated with Impacket utility suite modules, specifically wmiexec.py, smbexec.py, and secretsdump.py. These tools are commonly used by adversaries for remote command execution, lateral movement, and dumping of sensitive domain or system credentials.
Detects attempts to bypass or tamper with the Antimalware Scan Interface (AMSI) in-memory by monitoring command lines of common script-hosting processes (e.g., PowerShell, WScript, Rundll32) for strings indicative of AmsiScanBuffer patching, reflection, or manual configuration of AMSI initialization states.
This rule detects potential command-and-control (C2) activity by correlating the creation of known Cobalt Strike or Sliver framework named pipes with subsequent repeated HTTP/S network beaconing from the same process or host. The rule identifies processes establishing suspicious named pipes, then looks for persistent network connections to standard web ports (80, 443) within a 30-minute window, flagging instances where significant beaconing count is observed.
Detects instances where PowerShell or pwsh processes are initiated directly by explorer.exe containing command-line arguments indicative of obfuscated or hidden execution, often used in ClickFix-style social engineering attacks where victims are coerced into pasting malicious commands into the Windows Run dialog.
Detects the abuse of Windows protocol handlers ('search-ms:') and WebDAV UNC paths ('@SSL\DavWWWRoot') launched from common web browsers or Windows shell processes. This technique, frequently observed in 'ClickFix' phishing campaigns, enables attackers to stage remote payloads like LNK, HTA, or scripts by leveraging native Windows functionality to resolve remote content as if it were a local resource, bypassing typical file download security controls.
This rule detects a multi-stage attack pattern involving potential social engineering via lookalike conferencing domains. It identifies users visiting suspicious domains resembling well-known conferencing platforms (e.g., Teams, Meet, Zoom), followed by modifications to the Windows RunMRU registry key and subsequent execution of suspicious commands (PowerShell, CMD, mshta) by explorer.exe within a short timeframe.
Detects instances where the Windows Script Host (wscript.exe) is used to launch node.exe with specific command-line arguments (e.g., executing index.js) or a specific binary name, which is indicative of malicious script-based execution chains often seen in RAT deployment.
Detects suspicious PowerShell process executions initiated by Windows Explorer that include command-line arguments typically used for downloading and executing content from the internet. This pattern is commonly associated with fileless malware execution or malicious script droppers.
Page 30 of 465