Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
5,876 detections
Filters
Last updated
All Time
Detection languages
2,951
1,378
313
289
239
Contributors
1,807
1,529
1,304
180
173
Categories
20,456
11,839
5,876
5,155
4,865
Platforms
3,744
1,857
1,076
782
160
Products / Services
2,238
1,434
1,273
746
582
MITRE Techniques
2,448
2,148
1,129
937
787
CVEs
19
15
9
7
5
IDS Classtypes
834
334
94
23
20
IDS Protocols
689
276
235
117
23
Detects outbound IPv4 and IPv6 connections from endpoints to public addresses on ports other than DNS (53), HTTP (80), HTTPS (443), 8000.
Command-and-control tools and tunnels often use uncommon ports to avoid web-focused filtering. Private, loopback, link-local, multicast and documentation ranges are excluded, and repeated connections are grouped by endpoint, process, remote IP and port. Custom apps and admin tools also use these ports, so triage by process, destination reputation and how rare the destination is.
Command-and-control tools and tunnels often use uncommon ports to avoid web-focused filtering. Private, loopback, link-local, multicast and documentation ranges are excluded, and repeated connections are grouped by endpoint, process, remote IP and port. Custom apps and admin tools also use these ports, so triage by process, destination reputation and how rare the destination is.
Detects indicators of compromise associated with a trojanized Claude Code VS Code extension distributed through Cursor's marketplace. The rule identifies file modifications to authorized_keys, process executions establishing reverse SSH tunnels, specific hardcoded C2 IP addresses, and references to the attacker's SSH key 'svc-telemetry'.
Detects SSH backdoor artifacts of a trojanized Claude Code editor extension: connections to the tunnel host, reverse SSH forwards to local port 22 and dropped persistence files. Needs Defender for Endpoint network, process and file events and covers only published indicators. The query ran without errors in a live tenant.
Detects inbound exploitation sources and outbound callback destinations associated with PHP2Shell infrastructure (CVE-2026-109618), correlating both directions per device while filtering out standard browser traffic.
Detects web servers, PHP processes, or shell processes spawning fetch utilities such as wget or curl to communicate with a specific malicious payload IP address (162.35.99.94), often indicative of web shell exploitation or post-exploitation file download activity.
Detects processes reconnecting to a public IPv4 or IPv6 address after at least 7 days of inactivity on ports other than DNS (53), HTTP (80), HTTPS (443), 8000.
Command-and-control implants often sit idle and then call back to infrastructure they have contacted before. Private, loopback, link-local, multicast and documentation ranges are excluded, and connections are grouped by endpoint, process, remote IP and port, then split into the last 24 hours and the prior history. Backup agents, license checks, update services and scheduled jobs also reconnect on a long cycle, so triage by process, destination reputation and how rare the destination is.
Run this over a search window of 30 days or more.
Command-and-control implants often sit idle and then call back to infrastructure they have contacted before. Private, loopback, link-local, multicast and documentation ranges are excluded, and connections are grouped by endpoint, process, remote IP and port, then split into the last 24 hours and the prior history. Backup agents, license checks, update services and scheduled jobs also reconnect on a long cycle, so triage by process, destination reputation and how rare the destination is.
Run this over a search window of 30 days or more.
Detects the execution of an application named avscan.exe from suspicious writable directories such as Users Public, ProgramData, or AppData Local, followed by an outbound public network connection within a one-hour window.
Detects reconnaissance commands (whoami, systeminfo) and PowerShell execution spawned as children of remote-access tools (such as AnyDesk, ScreenConnect, TeamViewer/QuickAssist variants, etc.), followed shortly by an outbound network connection to a .top domain from the same device.
Detects a sequence of events on Linux hosts involving a remote interactive or network logon from an external IP, followed within 5 minutes by root-level process execution, and subsequently within 2 hours by an outbound SSH connection to a private or known attacker IP. Also includes a separate branch for direct logons from known malicious IPs.
Detects the execution of the rclone.exe utility with copy, sync, move, or copyto command-line arguments, correlated with outbound network connections to Wasabi S3 or MEGA cloud storage domains.
Detects process command lines containing Magento database audit strings (cms_block, cms_page) alongside specific loader domains, or network events communicating with these known loader domains.
Detects msiexec.exe spawning scripting or programming engines (such as PowerShell or Python) that perform LDAP directory searches and establish outbound external network connections within a specific time window.
Detects command-line execution of script interpreters (PowerShell, CMD, MSHTA) spawned via Windows Explorer containing download primitives and URLs, followed by immediate external network connections.
Detects instances where an AOMEI-related process initiates an outbound WebDAV-style connection to a public host, followed closely by a Windows WebClient service (svchost.exe) connection to the same remote IP address.
Detects when a Java or JavaW process spawns a Unix shell (bash, sh, dash) with command-line arguments indicative of a reverse shell utilizing /dev/tcp and interactive shell flags.
Detects processes that make outbound API calls to known Large Language Model (LLM) domains while simultaneously performing high-volume network connection attempts or scanning-like fan-out activity across multiple public IP addresses.
Detects outbound IPv4 and IPv6 connections from endpoints to public IP addresses on high-numbered ports (1023 and above), excluding private, reserved IP ranges and known noisy processes. Command-and-control tools, malware and tunneling activities may use uncommon ports to evade network monitoring.
Connections are aggregated by endpoint, process, remote IP and port to identify suspicious communication patterns.
Legitimate applications, custom services and administrative tools may also use these ports, so triage based on process context, destination reputation and connection frequency.
Connections are aggregated by endpoint, process, remote IP and port to identify suspicious communication patterns.
Legitimate applications, custom services and administrative tools may also use these ports, so triage based on process context, destination reputation and connection frequency.
Detects when Node.js or NPM processes access sensitive credential files such as .npmrc, SSH private keys, or AWS credentials, followed by an outbound public network connection from the same device within a 15-minute window.
Detects the installation or service setup of unauthorized remote access tools (such as AnyDesk, TeamViewer, or ScreenConnect) followed by outbound public network connections from the same device within a one-hour window.
Detects outbound network connections from Python processes associated with known indicators of compromise or specific suspicious command line arguments such as cook.py, cook.pth, or privcollect.
Detects network connections to the npm registry (registry.npmjs.org) initiated by AI coding agent processes (such as Claude, Cursor, Copilot, Windsurf, Codeium, Aider) or by unexpected processes that are not standard package managers or browsers.
Page 1 of 280