Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,084 detections
Filters
Last updated
All Time
Detection languages
418
193
163
156
81
Contributors
381
251
92
71
40
Categories
498
271
255
135
123
Platforms
443
441
267
121
114
Products / Services
10,665
9,758
6,692
4,418
3,749
1,084
MITRE Techniques
278
229
180
177
151
CVEs
17
15
13
13
8
IDS Classtypes
4
IDS Protocols
4
This rule detects a multi-stage attack chain targeting a single user within a 2-hour window. It identifies the combination of email bombing (a distraction technique), an incoming external or anonymous Teams call (social engineering channel), and the installation of common remote access tools (e.g., AnyDesk, TeamViewer) on the same host.
Detects the creation of a Git branch where the branch name matches the pattern of a 40-character hexadecimal SHA-1 hash. This behavior is associated with the 'Plugin4Shell' technique, which exploits how some git implementations prioritize matching ref names over target commit objects, enabling an attacker to manipulate or swap code in automated systems or AI coding agents that rely on SHA-pinning.
Detects potential indirect prompt injection attacks against LLM agents. The rule monitors for ingested or retrieved content containing known injection patterns (e.g., hidden comments, 'system:' role spoofing) that is followed within 5 minutes by a privileged tool invocation by the agent, such as data exfiltration or system command execution.
Detects anomalous AI orchestration or API gateway activity that suggests potential command and control (C2) communication. The rule monitors for unauthorized external endpoints, high-frequency low-payload beaconing patterns, and unauthorized cross-tenant agent-to-agent calls.
Detects anomalous spikes in AI/ML API usage for specific callers, including request volume, token count, and estimated costs, relative to a 7-day historical baseline. This activity is indicative of Cost Harvesting or Resource Exhaustion against pay-per-use AI services, mapped to the MITRE ATLAS framework.
Detects anomalous AI agent behavior characterized by excessive recursive tool-call chain depth or task durations exceeding the 95th percentile baseline. This activity is indicative of adversarial prompting designed to increase operational compute costs or induce resource exhaustion.
Detects anomalous query patterns against an AI inference API, characteristic of a membership inference attack. The rule identifies callers submitting a high volume of near-duplicate or minimally-perturbed inputs while requesting raw confidence scores, suggesting an attempt to probe the model to determine if specific data points were used in its training set.
Detects anomalous spikes in inference API request volume per caller, correlated with increased p95 latency and elevated error rates (5xx or 408). This behavior is characteristic of a denial-of-service attack targeting AI inference endpoints, potentially exhausting backend resources and disrupting service availability.
Detects sustained, statistically significant degradation in AI model performance (confidence and accuracy) and anomalous drift scores associated with specific callers or input sources. This behavior suggests a targeted attempt to erode model integrity through adversarial inputs rather than organic data drift.
Detects attempts to perform prompt injection or meta-prompt probing against an LLM gateway to extract confidential system instructions, as well as instances where the LLM response echoes sensitive system-level boilerplate consistent with a successful prompt extraction attack.
Detects attempts to perform prompt injection or meta-prompt probing against an LLM gateway to extract confidential system instructions, as well as instances where the LLM response echoes sensitive system-level boilerplate consistent with a successful prompt extraction attack.
Detects anomalous high-frequency requests to machine learning model endpoints from a single client, a pattern indicative of model extraction (stealing) attacks. This rule monitors inference request paths and success statuses to identify potential systematic input-space probing.
Detects adversarial attempts to poison Retrieval-Augmented Generation (RAG) knowledge bases. This includes the ingestion of documents containing prompt injection directives (e.g., 'ignore previous instructions'), hidden malicious markup intended to manipulate LLM inference, unauthenticated data ingestion, or abnormal volume spikes from a single source.
Detects attempts to evade machine learning model classifications by crafting adversarial perturbations against an inference endpoint. The rule identifies suspicious patterns, including a high volume of near-identical inputs from a single client resulting in confidence scores near the model's decision boundary, which is indicative of an automated optimization loop searching for an adversarial evasion attack (MITRE ATLAS T0043).
Detects Large Language Model (LLM) responses that contain sensitive information such as PII, internal proprietary markers, or credentials. This rule monitors model output for patterns matching social security numbers, credit card numbers, private keys, API keys, and internal company confidentiality labels, which could indicate data leakage or unintended disclosure.
Detects anomalous high-volume API requests from a single user/key to AI model inference endpoints. The rule tracks query diversity and total volume within an hour to identify behavior consistent with model extraction or cloning (AML.T0024.002) via systematic input/output pair collection.
Detects anomalous high-frequency queries to machine learning model inference endpoints involving probability or confidence score outputs. This behavior is indicative of a membership inference attack, where an adversary attempts to determine if specific data samples were included in a model's training set by harvesting confidence scores.
Detects high-frequency, sustained queries to an ML inference endpoint from a single identity, indicative of gradient-free training data reconstruction or model inversion attacks. The rule triggers when a specific user or API key performs over 100 successful requests within a 15-minute window, where the responses include granular confidence, probability, or logit values, which are typical targets for reconstructing training data.
Detects potential model weight poisoning or unauthorized integrity compromise by identifying training jobs launched by non-authorized users, model file modifications in production paths that bypass CI/CD approval gates, or integrity mismatches between deployed model artifacts and their corresponding approved training outputs.
Detects anomalous high-frequency queries to machine learning model inference endpoints involving probability or confidence score outputs. This behavior is indicative of a membership inference attack, where an adversary attempts to determine if specific data samples were included in a model's training set by harvesting confidence scores.
Detects high-frequency, sustained queries to an ML inference endpoint from a single identity, indicative of gradient-free training data reconstruction or model inversion attacks. The rule triggers when a specific user or API key performs over 100 successful requests within a 15-minute window, where the responses include granular confidence, probability, or logit values, which are typical targets for reconstructing training data.
Page 17 of 52