Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
7,138 detections
Filters
Last updated
All Time
Detection languages
3,224
1,086
498
476
469
Contributors
1,824
1,607
872
505
300
Categories
2,976
2,470
1,061
1,024
908
Platforms
39,790
7,138
6,515
3,861
3,597
Products / Services
1,325
1,067
970
655
490
MITRE Techniques
2,227
1,319
1,319
874
873
CVEs
66
65
33
32
28
IDS Classtypes
81
79
56
27
10
IDS Protocols
149
98
6
5
4
Experimental behavioral hunt for a root-owned shell or interpreter started by a non-root initiating account on explicitly inventoried ISE devices. This identifies an account-context transition in process telemetry; it does not establish preceding exploitation, setuid abuse, or a specific Cisco CVE. Requires verified appliance process telemetry.
Detects the registration or installation of an MCP (Model Context Protocol) server or agent tool containing suspicious prompt-injection strings (e.g., instructions to ignore safety guardrails or role overrides), immediately followed by the execution of shell commands or unauthorized network connections originating from that tool.
Detects unauthorized modifications to AI agent configuration files, including system prompts, tool allow-lists, and MCP server registration files. Such changes are often performed by processes outside of approved configuration management or infrastructure-as-code pipelines, potentially allowing for persistent manipulation of AI model behavior and capabilities.
Detects outbound HTTPS connections to major generative AI inference endpoints originating from non-browser, non-development tool processes at regular intervals. This activity is indicative of beaconing behavior, where malware (specifically families like LAMEHUG) leverages LLM APIs for dynamic command generation, payload obfuscation, or C2 instruction retrieval, bypassing traditional security controls.
Detects a two-stage activity: first, the potential unauthorized reading or access of LLM provider API keys from files, environment variables, or command-line arguments; and second, the subsequent use of those same credentials to authenticate to LLM provider APIs (OpenAI, Anthropic, Azure, Bedrock) from a different, unrecognized source IP address, indicating potential credential theft and session hijacking.
Detects the presence of package.json or node_modules artifacts associated with a known typosquatting campaign targeting the 'sorted-btree' npm package. The rule identifies suspicious package names or dependencies by matching against a list of known malicious packages used in the campaign, or by detecting the co-occurrence of the malicious 'indexed-btree' package alongside the legitimate 'sorted-btree' package.
Detects exploitation attempts by the CARBONATO botnet targeting exposed Docker daemons. The rule monitors for container escape techniques using 'nsenter' to access host namespaces, the execution of specific malicious entrypoint scripts, the use of known malicious container image references, and the creation of privileged containers with host bind mounts, which are indicative of host filesystem and network compromise.
Detects non-interactive 'sudo -n' used to test for or invoke a passwordless sudo rule
(e.g. 'sudo -n true', 'sudo -n -l', 'sudo -n bash'). Directly follows the sudoers write in
the OVSwrap (CVE-2026-64531) chain and is a common post-exploit validation step.
Detection logic adapted from Elastic Security Labs - Linux LPE detection framework.
(e.g. 'sudo -n true', 'sudo -n -l', 'sudo -n bash'). Directly follows the sudoers write in
the OVSwrap (CVE-2026-64531) chain and is a common post-exploit validation step.
Detection logic adapted from Elastic Security Labs - Linux LPE detection framework.
Detects the Sckit supply chain implant reading sensitive credential files from a user's home directory. The rule monitors processes spawned by Node.js or Python interpreters (the primary loaders for the Sckit worm) attempting to access high-value files like SSH keys, .npmrc, .pypirc, and various token storage files, indicating potential credential harvesting behavior.
Detects potential runtime polymorphic malware that modifies its own executable or script content while simultaneously communicating with generative AI or LLM API endpoints. This behavior indicates an adversary using LLMs to regenerate obfuscated code or mutate malware signatures dynamically at runtime.
Detects repeated attempts to enumerate available model endpoints on a LiteLLM server, a behavior often associated with reconnaissance targeting AI gateway services to identify exposed models.
This rule detects the 'ClickFix' attack pattern, where an adversary tricks a user into copying and pasting a malicious command into a terminal. The command typically involves using 'base64 -D' to decode an attacker-supplied blob directly into a bash interpreter, often initiated by a Terminal emulator and containing strings related to fake reCAPTCHA or robot verification.
This rule monitors network connections, file events, and process execution command lines for references to a list of known malicious URLs, including indicators associated with ClearFake, IClickFix, AMOS, Remus, and Mozi botnet payloads.
Detects anomalous behavioral patterns where an AI agent process (e.g., Claude Code, Cursor, Aider) performs multi-category discovery of its own runtime environment, plugins, and host configuration. The rule identifies agents that trigger processes spanning at least three distinct discovery categories (Network, File/Plugin, Software, or System) within a short timeframe, indicating potential reconnaissance of agent-authorized capabilities and tools rather than standard host exploration. Covers T1082, T1518, T1083, T1016
Detects incoming HTTP POST requests to the 'diffpatch' endpoint in Gitea that contain suspicious 'diff --git' patterns combined with path traversal sequences ('../') targeting 'hooks/' directories. This indicates an attempt to exploit CVE-2026-60004 to achieve unauthorized file system access or remote code execution via repository hooks.
TokenGrabber Builder: Nuitka/PyInstaller AV-Evasion Compilation - detects Python builder scripts compiling an embedded stealer.py payload into a native executable via Nuitka or PyInstaller with AV-evasion flags
Detects the use of package installation commands (pip, npm, npx) where the package name resembles popular AI/ML libraries (tensorflow, torch, openai, langchain) but is not an exact match for known legitimate versions. This is a common indicator of a typosquatting supply chain attack, where attackers attempt to trick users into installing malicious packages with similar names to trusted software.
This rule detects the use of headless or automated web browser tooling (such as Selenium, Puppeteer, Playwright, or browser-specific drivers like chromedriver) establishing network connections to popular generative AI websites. This behavior can indicate automated interaction with AI services, potential data exfiltration of internal information, or unauthorized automation.
Detects the use of package installation commands (pip, npm, npx) where the package name resembles popular AI/ML libraries (tensorflow, torch, openai, langchain) but is not an exact match for known legitimate versions. This is a common indicator of a typosquatting supply chain attack, where attackers attempt to trick users into installing malicious packages with similar names to trusted software.
Detects suspicious command line activity or network interactions targeting MongoDB, characterized by the use of 'eval' combined with specific function calls (such as 'process', 'load', 'require', or constructor-chaining) indicative of attempts to escape the MongoDB JavaScript sandbox.
Detects HTML files that employ FlipBook branding as a lure for password-gated smuggling. The detection identifies client-side JavaScript logic that uses PBKDF2 and AES-GCM to decrypt a hidden redirect URL after user interaction with a password prompt.
Page 31 of 340