Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
9,763 detections
Filters
Last updated
All Time
Detection languages
3,724
2,040
1,335
1,055
969
Contributors
2,963
2,050
1,372
1,230
485
Categories
5,722
3,079
1,138
900
659
Platforms
9,650
658
432
53
30
Products / Services
10,703
9,763
6,706
4,421
3,752
MITRE Techniques
3,828
3,377
1,982
1,618
1,042
CVEs
18
14
12
12
11
Detects the execution of the Program Compatibility Assistant (pcalua.exe) with the '-a' flag to launch executables located in common user-writable directories such as Temp, Downloads, or AppData. This behavior is a known Living off the Land Binaries and Scripts (LOLBAS) technique used to proxy execution and potentially bypass security controls or break parent-child process lineage.
Detects the execution of wscript.exe or cscript.exe with VBScript or JScript files from user-writable and temporary directories, such as %TEMP%, %APPDATA%, and Downloads folders. This pattern is commonly associated with the execution of malicious droppers and loaders in malspam campaigns.
This rule detects the execution of the Windows FTP client (ftp.exe) using a command script file (via the -s: parameter). It monitors for process creation of ftp.exe with this flag, excluding known legitimate paths. It then correlates this with any child processes initiated by the FTP client to identify potential data exfiltration or automated file transfers.
Detects the installation of a GitLab Runner service on Windows, specifically identifying cases where the service is configured to run under the LocalSystem account. This is monitored via Service Control Manager (Event ID 7045), Sysmon (Registry Event ID 13), and process creation events involving sc.exe or gitlab-runner.exe.
This rule detects instances where common system utilities like notepad.exe or calc.exe are spawned as child processes by PowerShell.exe, specifically when the initiating PowerShell process was spawned by gitlab-runner.exe but the parent process information no longer resolves correctly. This is often indicative of persistence or post-exploitation activities occurring after a CI/CD job has ostensibly finished.
Detects a multi-stage, automated attack chain performed by AI-based development agents (e.g., Claude, Copilot, Cursor). The rule correlates process discovery, credential access, file collection (staging), and exfiltration over web services within a tight 15-minute window, identifying potential autonomous exploitation and data exfiltration. Covered T1046, T1552.001, T1078, T1005, T1567
This rule detects scenarios where an AI agent executes shell commands or system-level tools immediately after receiving a prompt injection attack from an untrusted source. It correlates agent audit logs—specifically identifying tool calls flagged for prompt injection—with actual endpoint process creation events. It flags process execution as 'High' severity if the command line matches common malicious patterns like encoded commands, IEX, or attempts to access credential material (e.g., lsass, mimikatz).
This rule detects scenarios where an AI agent executes shell commands or system-level tools immediately after receiving a prompt injection attack from an untrusted source. It correlates agent audit logs—specifically identifying tool calls flagged for prompt injection—with actual endpoint process creation events. It flags process execution as 'High' severity if the command line matches common malicious patterns like encoded commands, IEX, or attempts to access credential material (e.g., lsass, mimikatz).
Detects the use of bcdedit.exe to configure the boot environment to start Windows in 'safeboot' mode with networking enabled. Adversaries may use this to bypass security controls that only run in normal boot mode or to facilitate persistent access with minimal system interference.
Detects memory regions associated with 'ntdll.dll' that are marked with Read, Write, and Execute (RWX) permissions. This pattern is commonly used in process injection techniques, such as reflective DLL injection or shellcode execution, where adversaries prepare executable memory space within a legitimate process to bypass standard security controls.
This rule detects the use of 'taskkill.exe' to terminate critical services, including security software (MsMpEng.exe), database processes (sql.exe, oracle.exe, sqlservr.exe), and common user applications (outlook.exe, onedrive.exe). This behavior is characteristic of the DragonForce ransomware, which disables protective services and applications before encryption to minimize interference and ensure successful file locking.
Detects high volumes of file rename operations within a short timeframe, which is a behavioral indicator often associated with the encryption process of ransomware such as CRPx0.
Detects the CRPx0 ransomware technique of unhooking ntdll.dll in memory. This is achieved by reading the ntdll.dll file from disk and overwriting the in-memory .text section with a clean, unhooked version to bypass EDR monitoring and execute direct syscalls.
Detects the execution of SoftPerfect Network Scanner (netscan.exe), a tool frequently utilized by threat actors such as Ransom Busters for internal network reconnaissance and host discovery activities prior to lateral movement or data exfiltration.
Detects the modification of boot configuration data (BCD) using bcdedit to enable 'Safe Mode with Networking' on the next reboot. Adversaries, including Akira ransomware affiliates, use this technique to bypass endpoint detection and response (EDR) solutions by ensuring they do not start when the system boots in a restricted safe mode.
Detects the creation of a local user account using the hardcoded 'Numlock!123' password, a characteristic artifact associated with Ransom Busters threat group operations. The rule monitors command-line activity from net.exe or net1.exe to identify attempts to add users with this specific, known malicious credential.
Detects the loading of known vulnerable kernel drivers 'truesight.sys' or 'rentdrv2.sys'. These drivers are frequently abused in Bring-Your-Own-Vulnerable-Driver (BYOVD) attack chains, where attackers leverage specific IOCTL calls (such as 0x22E044 or 0x22E010) to interact with the driver to perform privileged actions, such as terminating security processes.
Detects the termination of critical processes, including security software (MsMpEng.exe), database services (sql.exe, oracle.exe, sqlservr.exe), and office productivity applications (excel.exe, outlook.exe, winword.exe). The sudden termination of these processes can indicate unauthorized attempts to disable security controls, disrupt database operations, or interfere with end-user activity.
Detects the use of administrative utilities 'wmic.exe' and 'schtasks.exe' to execute commands or create scheduled tasks on remote systems, excluding local execution scenarios.
Detects the loading of rstrtmgr.dll by processes other than explorer.exe. Adversaries, particularly ransomware, utilize the Windows Restart Manager API (RmStartSession, RmRegisterResources, RmShutdown) to identify and forcibly terminate processes that hold file handles to files they intend to encrypt, thereby bypassing file-in-use locks.
Detects the use of bcdedit.exe to modify boot configuration data to enable Safe Mode with Networking. This technique is used by Akira ransomware affiliates to potentially disable or impair EDR and other security tooling prior to deploying the ransomware payload.
Page 85 of 465