Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
7,144 detections
Filters
Last updated
All Time
Detection languages
3,228
1,086
498
478
469
Contributors
1,824
1,607
872
505
300
Categories
2,979
2,473
1,063
1,026
910
Platforms
39,802
7,144
6,519
3,861
3,601
Products / Services
1,325
1,070
970
655
490
MITRE Techniques
2,231
1,319
1,319
876
873
CVEs
66
65
33
32
28
IDS Classtypes
81
80
56
27
10
IDS Protocols
149
98
6
5
4
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 usage of the Android Debug Bridge (ADB) 'pair' command initiated by processes other than known legitimate Android development tools (e.g., adb.exe, Android Studio). This may indicate an attacker attempting to wirelessly pair a malicious device or gain unauthorized access to an Android device over the network.
This rule detects command-line activity where a remote resource is downloaded via 'curl', decoded using 'base64', and then piped directly into a shell interpreter (sh, bash, or zsh). This pattern is commonly used in fileless execution or stage-one payload delivery by attackers to execute malicious scripts directly from memory.
Detects the execution of the s5cmd utility using 'cp' or 'sync' commands directed towards S3 storage paths. This behavior is indicative of unauthorized data exfiltration to attacker-controlled cloud storage, a tactic identified in intrusions associated with the 'Ransom Busters' group.
This rule monitors network and event logs to detect instances where a single user account, logged into a specific host, initiates connections to two or more distinct, potentially suspicious remote IP addresses. This pattern of multi-destination connection activity may indicate command and control communication, scanning behavior, or data staging.
Detects anomalous Telnet (TCP port 23) traffic associated with Mirai-variant botnet propagation, such as scanning for new targets or inbound compromise attempts, often following cPanel/WHM vulnerabilities.
Detects the execution of 'bun' during a Node.js package installation process (npm or node). This behavior can be indicative of a supply chain attack where malicious actors attempt to leverage alternative runtimes during the 'preinstall' phase or package installation to execute arbitrary code or bypass security controls typically associated with standard Node.js installations.
Detects malformed HEIF/HEIC image files with anomalous ftyp/box structure consistent with the libheif memory corruption exploit (CVE-2026-32882) used against Discourse forum image upload endpoints
Detects malformed HEIF/HEIC image files with anomalous ftyp/box structure consistent with the libheif memory corruption exploit (CVE-2026-32882) used against Discourse forum image upload endpoints
Identifies Debian systems running vulnerable versions of the libheif package (versions 1.10.x through 1.21.x) that are susceptible to CVE-2026-32882. This rule uses software inventory data to detect installations requiring an upgrade to version 1.22.0 or higher.
Identifies Debian systems running vulnerable versions of the libheif package (versions 1.10.x through 1.21.x) that are susceptible to CVE-2026-32882. This rule uses software inventory data to detect installations requiring an upgrade to version 1.22.0 or higher.
Detects web server processes (e.g., Apache, Nginx, PHP-FPM, IIS) launching suspicious shell or system administration utilities. This behavior is indicative of a web server compromise, potentially involving the use of web shells or the exploitation of public-facing web applications to execute arbitrary commands.
This rule detects the presence of specific SilverFox malware payload files by matching their SHA256 hashes against known indicators of compromise. This identification is typically used for file integrity monitoring, endpoint scanning, or gateway-based file filtering to prevent malicious code execution.
Detects the presence of specific versions (2.35 and 2.36) of the Admin Menu Editor Pro WordPress plugin, which were reported to be distributed as trojanized versions via a compromised update infrastructure.
Detects high-frequency outbound TCP connection attempts from a WSO2 API Gateway or Traffic Manager host to multiple sensitive internal management and administrative ports (22, 445, 3389, 5985, 5986). This behavior is indicative of a compromised gateway being used for lateral movement across the network.
Detects local privilege escalation attempts targeting Acronis backup plugins for cPanel, WHM, and Plesk. The rule monitors for instances where backup-related processes or binaries located within control panel paths are executed with root privileges when the parent process was not originated by root, a behavior indicative of exploiting insecure file permissions.
Detects the publication of a known malicious version (0.2.21) of the '@dforge-core/dforge-mcp' npm package. This activity is attributed to the GHAPPIER/PolinRider supply chain campaign, which leverages hijacked OIDC trusted-publishing pipelines to distribute compromised code.
This rule monitors network traffic, DNS queries, and user web clicks to detect interactions with known malicious domains associated with credential harvesting and phishing campaigns, specifically those impersonating security or SSO portals.
Detects instances where AI-based coding or agent-driven tools (e.g., Cursor, Claude Code, AutoGPT) perform suspicious file archiving/packaging activities followed by network egress to unauthorized or known exfiltration-prone services. This helps identify potential data exfiltration by AI agents or compromised development environments.
Detects instances where AI-based coding or agent-driven tools (e.g., Cursor, Claude Code, AutoGPT) perform suspicious file archiving/packaging activities followed by network egress to unauthorized or known exfiltration-prone services. This helps identify potential data exfiltration by AI agents or compromised development environments.
Page 65 of 341