Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
7,139 detections
Filters
Last updated
All Time
Detection languages
3,217
1,086
504
478
469
Contributors
1,824
1,607
872
505
300
Categories
2,971
2,471
1,060
1,026
908
Platforms
39,762
7,139
6,517
3,851
3,599
Products / Services
1,324
1,067
970
657
487
MITRE Techniques
2,228
1,317
1,314
873
871
CVEs
66
65
33
32
28
IDS Classtypes
81
80
56
27
10
IDS Protocols
149
98
6
5
4
Detects malicious command injection where GitHub Actions interpolates ${{ github.event.comment.body }} directly into a bash run block, specifically looking for patterns associated with the 'curl.litter.catbox.moe..sh' domain.
Detects vLLM processes spawning interactive Unix shells, a strong indicator of remote code execution, potentially exploiting CVE-2026-22778.
Detects vLLM executing inline Python commands using the '-c' flag, a common post-exploitation technique. This rule specifically looks for process command lines containing 'vllm' and a Python interpreter (python or python3) executing code via the '-c' flag, which is indicative of inline command execution.
Correlates the creation and immediate deletion of the trin.tar.gz archive handled by the Python temporary-directory context manager.
Correlates local LLM execution engines (Ollama, LM Studio) unexpectedly spawning command shells, a strong indicator of successful prompt execution payloads.
Detects the creation of '.pkg_history' or '.pkg_logs' files in the OS temporary directory by a Node.js process. This behavior is linked to the easy-day-js supply chain dropper, where the '.pkg_logs' file contains XOR-encoded campaign marker bytes.
Detects the execution of processes associated with the malicious npm package 'apintergrationpost', which acts as a masquerade for a Linux RAT (Remote Access Trojan) labeled as MYRA. The rule monitors process launch events for command lines or file paths containing the package name.
Detects the execution of a suspicious 'setup.cjs' script via Node.js triggered during an 'npm install' process, characteristic of the malicious payload delivery mechanism found in compromised Mastra packages. The rule specifically monitors for Node.js spawning a process with command-line arguments including 'setup.cjs' and the '--no-warnings' flag.
Detects the creation of a 'protocal.cjs' file within the '~/.config/systemd/nvmconf/' directory. This activity is associated with the 'easy-day-js' supply chain attack, where a malicious Node.js script is established as a persistent user-level systemd service named 'nvmconf'.
Detects instances where a process associated with the 'brand-landingpage' AI agent or processes involving scripts from 'stitch-design.ai' spawn child processes, potentially indicating malicious activity leveraging the agent's runtime permissions.
Detects the creation of a specific ransom note file associated with ShinyHunters activity within PeopleSoft web or application server configuration directories.
Detects the deployment of MeshCentral remote management agents and associated outbound communication to suspicious C2 infrastructure, identified in environments compromised by ShinyHunters. This rule monitors for specific process execution patterns involving meshctrl.js and meshagent, alongside network connections to known C2 domains and IP addresses.
Detects the use of archive utilities (tar, zip, 7z, gzip, zstd) to bundle sensitive PeopleSoft configuration and application directories (e.g., ps_config_homes, psoft/psft). This behavior is consistent with data staging for exfiltration, frequently associated with the ShinyHunters threat group activity targeting exposed PeopleSoft environments.
Detects the use of the zstd compression utility on PeopleSoft application servers by privileged service accounts such as psoft, oracle, or linuxadm. This behavior is indicative of data staging or archiving activity consistent with ShinyHunters pre-exfiltration tactics.
Udm Ghost Dynamic Js
YARA-L
Targets JavaScript containing atob decoding, dynamic appendChild, and one-time localStorage keys. The UDM rule retains principal and target context for correlation.
Detects a multi-stage npm supply chain compromise, often associated with Operation Friday Harvest or PostCSS typosquatting campaigns. The rule monitors for a sequence of 4 out of 5 behavioral stages on a single host within a 30-minute window: malicious postinstall script execution (npm/node spawning shells), staging of malicious payloads in temp directories, network connections to known command-and-control (C2) or payload-hosting infrastructure (e.g., GitHub, Telegram), execution of dropped binaries, and unauthorized access to sensitive credential files.
This rule detects potential supply chain attacks where a Node.js process (npm, npx, node) triggers an unauthorized shell, scripting interpreter, or network utility (such as curl, wget, powershell, or bash) to execute commands. This behavior, when followed by file creation in temporary directories (e.g., /tmp, AppData\Local\Temp) within a 5-minute window, is indicative of malicious npm postinstall hooks, commonly seen in supply chain compromise campaigns.
Detects unauthorized file read operations targeting local cryptocurrency wallet directories and browser extension storage (MetaMask, Phantom, etc.) by suspicious processes (e.g., node.exe, chost.exe) or processes executing from temporary or user-writable application directories, commonly associated with infostealer malware.
Detects unauthorized access to cloud provider credential files (AWS, GCP, Azure) by potentially suspicious processes. It specifically monitors processes such as node.exe, npm, python, bash, or binaries running from temporary directories (/tmp, %TEMP%) while excluding legitimate CLI tools.
Identifies execution of the Linux/macOS bash cradle used by GLASSWORM's WASM loader. It looks for curl fetching the specific OS path structure ('/linux/i/_' or '/darwin/i/_') and piping to bash.
Identifies potential Neo-reGeorg webshell execution by correlating HTTP POST requests to Java Server Pages (.jsp/.jspx) with reflection events (defineClass) in system logs. This specific mechanism is utilized by Operation Escaneo (MexicanMafia) to load AES-encrypted, base64/BLV-encoded payloads in memory.
Page 244 of 340