Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
5,156 detections
Filters
Last updated
All Time
Detection languages
2,302
1,070
640
289
288
Contributors
918
814
769
696
224
Categories
20,456
11,839
5,877
5,156
4,865
Platforms
2,859
2,017
816
524
213
Products / Services
1,337
1,144
831
576
366
MITRE Techniques
1,823
1,307
787
747
577
CVEs
9
7
7
6
6
IDS Classtypes
538
151
101
57
48
IDS Protocols
375
373
132
64
42
Detects DeepLoad-related behavioral chains including ClickFix PowerShell execution, mshta retrieval, scheduled task persistence, Add-Type or APC style injection, LockAppHost or related target process launches, WMI based re-execution, filemanager.exe execution, USB spread indicators, and known DeepLoad infrastructure communication.
This rule detects unusual network connections from processes to legitimate service endpoints that are commonly abused for Command and Control (C2) activities (LOLC2). It specifically looks for connections to services like Telegram, Slack, Discord, GitHub, Google APIs, Pastebin, Dropbox, and others, originating from processes that are not recognized as legitimate applications for those services, and not from common system processes or standard program file paths. This helps identify potential adversary communication channels masquerading as legitimate traffic.
This rule detects various stages of DeepLoad malware activity, including initial delivery via PowerShell execution policy bypass with IEX/IRM style ClickFix, network retrieval using mshta.exe, persistence mechanisms like scheduled task creation and WMI, fileless injection through PowerShell Add-Type compilation and suspicious child processes (LockAppHost.exe, makecab.exe, Magnify.exe), credential stealing via filemanager.exe, and USB propagation indicators including specific C2 paths and LNK themed installer names. It also identifies known DeepLoad infrastructure domains.
Detects suspicious execution of rundll32.exe initiated by explorer.exe that includes network-related command-line arguments (http, https, .dll, entrypoint). This pattern can indicate malicious DLL loading or execution attempting to establish network communication, potentially for command and control or data exfiltration. The rule filters out common system accounts and focuses on infrequent occurrences.
Detects potential NetScaler administrative session hijacking by monitoring for the usage of SESSID or NITRO_SK administrative cookies from multiple source IPs. Attackers leverage CVE-2026-3055 to steal these specific session cookies to gain unauthorized administrative access.
This rule detects various indicators of compromise associated with the GlassWorm malware. It monitors for network connections to known GlassWorm remote IPs and domains, process activity involving specific GlassWorm filenames, folder paths, command line arguments (including wallet addresses and command terms), and registry modifications for persistence (Run/RunOnce keys with GlassWorm-related values). The rule correlates multiple evidence types to reduce false positives.
This detection rule identifies network activity associated with the Torg Grabber infostealer. It specifically looks for connections to known command and control (C2) domains and specific URL paths used by the malware for authentication, data upload, or core functionality. The rule leverages network proxy, firewall, or Zeek HTTP logs to extract and analyze URLs. A match indicates a strong reason to investigate the affected host and user, check for expected access, and review nearby activity for further compromise.
Detects instances where rundll32.exe is launched by explorer.exe and includes command-line parameters indicative of network communication (http, https) or WebDAV activity (davclnt.dll, WebDAV, \\). This behavior can be associated with forced authentication attempts or execution of remote code.
Detects the execution of S3 Browser or Office installation executables, followed by network traffic from the same host to known cloud storage or file-sharing sites within a short time window. This behavior aligns with Muddled Libra's method of staging and exfiltrating sensitive data, such as PST files, to attacker-controlled cloud buckets.
Detects network communication patterns associated with EtherRAT, including beaconing to 'aurineuroth.com' and CDN-like URLs. Additionally, it identifies host-based fingerprinting activities often performed by EtherRAT, such as hidden PowerShell execution, discovery of MachineGuid, antivirus product information, domain membership, and video controller details, as well as the creation of 'svchost.log' in AppData.
This rule detects unusual network connections from processes to legitimate service endpoints that are commonly abused for Command and Control (C2) activities (LOLC2). It specifically looks for connections to services like Telegram, Slack, Discord, GitHub, Google APIs, Pastebin, Dropbox, and others, originating from processes that are not recognized as legitimate applications for those services, and not from common system processes or standard program file paths. This helps identify potential adversary communication channels masquerading as legitimate traffic.
This rule detects the presence of specific 'NODE_CHANNEL' values ('anglmf', 'cryptoexth4', 'complexarchaeologist1') in process command lines or network event initiating process command lines on macOS systems. These values have been observed in the GhostClaw/GhostLoader malware campaign, indicating potential execution or network communication related to this threat.
This rule detects the use of common utilities to transfer executable files (shell scripts, python scripts, perl scripts, or ELF binaries) between systems.
This rule detects activities associated with the 'CTRL' Russian .Net access framework. It looks for a combination of a hidden PowerShell stager, UAC bypass attempts via registry modification (ms-settings), and the launch of fodhelper.exe. Additionally, it requires the presence of at least one of the following: disguised registry payload storage, FRP configuration file staging, keylogger artifact creation, named pipe creation ('ctrlPipe'), RDP tampering artifacts, or PowerShell commands related to Windows Defender exclusion changes.
Detects Node.js making outbound connections to suspicious domains or IPs
This detection rule identifies activity related to the TeamPCP Telnyx PyPI backdoor, specifically targeting versions 4.87.1 and 4.87.2 of the 'telnyx' package. It looks for command-line executions involving 'pip install telnyx==4.87.1' or 'pip install telnyx==4.87.2'. Additionally, it detects network connections to the IP address 83.142.209.203 on port 8080 requesting 'ringtone.wav' or 'hangup.wav', which are indicators of the backdoor's communication. The rule also identifies persistence mechanisms by looking for 'msbuild.exe' in the Windows Startup folder path '\Microsoft\Windows\Start Menu\Programs\Startup\msbuild.exe'.
This detection identifies network connections to specific malicious domains associated with the TeamPCP LiteLLM PyPI compromise (litellm 1.82.7 / 1.82.8). It specifically looks for connections originating from Python-related processes (python, python3, pip) to domains like 'models.litellm.cloud' or 'checkmarx.zone'. A match indicates potential compromise and warrants investigation into the affected host, user, and surrounding activity.
This rule detects activity associated with the TeamPCP Telnyx PyPI backdoor, specifically targeting versions 4.87.1 and 4.87.2 of the 'telnyx' package. It identifies two main indicators: first, process command lines that show the installation of these specific malicious versions via 'pip' or 'python -m pip'; second, network connections to a known command and control (C2) IP address (83.142.209.203) or specific C2 URLs involving '.wav' files, which are known to be used for steganography in this campaign. A match indicates a high confidence of potential compromise and warrants immediate investigation to confirm the affected host and user, assess the legitimacy of the activity, and review for follow-on malicious actions.
KQL Query
This rule detects network connections and DNS resolution requests to known typosquatted domains used by the TeamPCP threat actor for command and control (C2) and data exfiltration. These domains are associated with a compromised Trivy scanner. Detection indicates successful execution of the compromised scanner and attempted exfiltration.
This rule detects rare external domains shared in Microsoft Teams messages that users subsequently click on. It identifies messages from external senders, extracts URLs and their domains, and then filters for domains that appear infrequently (less than 3 times in the last day). The rule then correlates these rare external domains with user click events within Teams, surfacing potential phishing attempts or malicious link sharing where users interact with untrusted content.
Page 185 of 246