Detections

Explore public detection logic contributed by the community across SIEM and rule languages.

7,276 detections

Detects when AI agent processes establish listening connections on network ports, indicating potential unauthorized service exposure or lateral movement activity.
avatar
Ankit Mehta@Secvyn
avatar
Detections.ai Community
5 months ago
6014
This rule broadly searches for the unique dead drop string 'OhNoWhatsGoingOnWithGitHub' across all command line executions, independent of node/bun, indicating potential compromise related to the Mini Shai-Hulud supply chain attack.
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
4 months ago
002
This rule detects the execution of Node.js processes that include specific attacker IP addresses (23.254.164.123:443 or 23.254.164.92:8000) in their command line arguments, combined with the creation of a file named '.pkg_logs'. This behavior is indicative of the Mastra NPM supply chain attack, where malicious packages drop a logging file after execution.
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
4 months ago
002
Detects the simultaneous creation of .pkg_history and .pkg_logs in a temporary directory, indicating the setup.cjs dropper associated with the Mastra NPM supply chain attack.
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
4 months ago
002
Looks for curl or wget downloading a payload followed by node or bun immediately executing a .js or .mjs file. This pattern is indicative of supply chain attacks, particularly those targeting JavaScript ecosystems like npm.
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
4 months ago
002
Identifies rapid dropping of many distinct files deep within the .npm/_npx/ directory structure, a precursor to worm execution. This rule specifically looks for more than 20 distinct file creations within a 5-minute window in paths matching the regex ".npm[\]_npx[\][0-9a-fA-F]+".
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
4 months ago
002
Detects the command line sequence used by Miasma to restore passwordless sudo for the 'runner' user in GitHub Actions by overwriting sudoers via a Docker container. This specific sequence involves echoing 'runner ALL=(ALL) NOPASSWD:ALL' into a file within the /mnt/runner directory and then changing its permissions to 0440, indicative of a privileged modification to the sudoers file within a containerized GitHub Actions runner environment.
avatar
Ethan Andrews@eandrews
avatar
Federal Signal Detections
5 months ago
11031
Detects PHP or web server processes launching command-line database clients. INFINITERED collected database credentials and supported arbitrary SQL execution against compromised REDCap environments. Direct database-client execution by a web-serving process is uncommon and may indicate credential access, data collection, or backdoor command execution.
avatar
Daniel Jeremiah@daniyyell
avatar
Detections.ai Community
4 months ago
002
Detects attempts to overwrite web server content files for defacement
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
105
Detects access to SSL/TLS private key files by unauthorized processes
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
105
Detects download of attacker tools via C2 ingress transfer channels
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
105
Detects use of trap command to handle system signals for Linux persistence
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
005
Detects the use of sshpass combined with disabled strict host key checking. This pattern matches the propagation scripts used by UNC6240 for rapid lateral movement across internal nodes following Oracle PeopleSoft exploitation.
avatar
Ethan Andrews@eandrews
avatar
Federal Signal Detections
5 months ago
4013
Detects high-volume use of AI APIs potentially for generating attack content
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
007
Detects unauthenticated access to Docker daemon API for container escape
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
004
Detects installation of rootkit kernel modules for deep system hiding
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
004
Detects bulk scraping of LinkedIn company profiles for employee intelligence
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
004
Detects suspicious chmod commands used to hide or protect malware
avatar
Darshan Thummar@Mighty
avatar
Detections.ai Community
5 months ago
004
This rule detects when package managers like npm, bun, or pnpm access SSH private key files. Such activity can indicate a supply chain attack where a malicious package attempts to exfiltrate sensitive credentials, granting unauthorized access to servers. If this alert fires, investigate the initiating process, check for newly installed software, and audit server logins associated with the compromised SSH key. Involve your security team if the cause is unclear.
avatar
Daniel Jeremiah@daniyyell
avatar
Detections.ai Community
5 months ago
108
This rule detects when a child process of npm (node, bun, or npm itself) reads a Kubernetes service account token file. This activity can indicate a compromised package attempting to access internal Kubernetes clusters from a developer's machine. Such an event warrants investigation into recent software installations or user activity, and engagement with the security team if the cause is unclear.
avatar
Daniel Jeremiah@daniyyell
avatar
Detections.ai Community
5 months ago
108
Detects the execution of the 'moladb_superusers.py' script, which is associated with the MuddyWater threat group for database enumeration.
avatar
Ibrahim Saud@tektrix
avatar
Detections.ai Community
5 months ago
003
Page 261 of 347