Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
60,142 detections
Filters
Last updated
All Time
Detection languages
23,200
16,896
4,280
4,077
3,289
Contributors
11,679
8,664
7,082
6,786
4,478
Categories
20,099
11,460
5,732
4,980
4,798
Platforms
39,723
6,855
6,349
4,079
3,510
Products / Services
10,351
9,601
6,991
4,335
3,859
MITRE Techniques
18,034
15,417
12,647
8,188
6,021
CVEs
68
68
58
56
50
IDS Classtypes
1,896
482
449
381
237
IDS Protocols
2,379
848
406
318
97
Detects evidence of indirect prompt injection attacks where an adversary introduces malicious instructions into data sources retrieved by an LLM or agent. The rule identifies common obfuscation and override techniques, including the use of zero-width characters, HTML comments for prompt breaking, base64 encoded instruction verbs, and specific override phrases within retrieved documents or tool outputs.
Detects anomalous data mutations in machine learning training pipelines, such as bulk insertions or label distribution shifts, performed by low-reputation or untrusted contributors. This activity is indicative of attempts to poison model training data to induce backdoors or skew decision boundaries.
This rule detects unauthorized tampering with AI container images within a container registry. It identifies when existing image tags are overwritten by actors outside of authorized CI/CD pipelines, or when production MLOps infrastructure pulls unsigned or invalid container images, potentially indicating a supply chain attack targeting AI model artifacts.
This rule monitors the behavior of AI agent orchestrators to detect potentially malicious activities, specifically focusing on unauthorized tool usage, deep tool invocation chains, and potential data exfiltration. It triggers when an agent invokes a tool outside of its authorized scope, performs an excessively deep sequence of tool calls, or accesses sensitive data followed by an outbound connection to an external endpoint, which is consistent with the MITRE ATLAS AML.T0053 technique.
Detects external reconnaissance activity targeting AI backend services. Adversaries scan for known AI infrastructure ports (such as TensorFlow Serving, Triton, or Ray) and common API paths (e.g., /v1/models, /predict) to identify potential targets. The detection utilizes proxy logs to correlate these suspicious network attributes, high request rates, or known scanning user agents.
Detects reconnaissance activities where users submit prompts containing keywords typically used to probe LLM internals, such as system prompts, special instructions, or configuration details. This behavior is often indicative of precursor activity for prompt injection or jailbreak attempts.
Detects attempts to extract or override the system instructions of an Large Language Model (LLM) via prompt injection techniques, such as explicitly asking for the system prompt or commanding the model to ignore previous instructions.
Detects ingestion of documents containing potential LLM prompt injection payloads (using zero-width characters or HTML comments to hide imperative instructions) into a RAG system, followed by anomalous LLM behavior such as unauthorized tool execution or unintended data disclosure.
Detects ingestion of documents containing potential LLM prompt injection payloads (using zero-width characters or HTML comments to hide imperative instructions) into a RAG system, followed by anomalous LLM behavior such as unauthorized tool execution or unintended data disclosure.
Detects ingestion of documents containing potential LLM prompt injection payloads (using zero-width characters or HTML comments to hide imperative instructions) into a RAG system, followed by anomalous LLM behavior such as unauthorized tool execution or unintended data disclosure.
Detects anomalous AI model inference activity where a caller submits a high volume of near-duplicate queries while frequently requesting raw confidence scores. This behavior is indicative of a membership-inference privacy attack, where an adversary attempts to determine if specific data points were included in the model's training set by analyzing query response patterns.
Detects LLM jailbreak attempts using a 'crescendo' multi-turn strategy. The rule monitors for sequences where an initial prompt contains known jailbreak-intent keywords leading to an AI refusal, followed shortly by a compliant AI response to a rephrased request within the same conversation session.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.
Detects LLM jailbreak attempts using a 'crescendo' multi-turn strategy. The rule monitors for sequences where an initial prompt contains known jailbreak-intent keywords leading to an AI refusal, followed shortly by a compliant AI response to a rephrased request within the same conversation session.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.
Detects LLM jailbreak attempts using a 'crescendo' multi-turn strategy. The rule monitors for sequences where an initial prompt contains known jailbreak-intent keywords leading to an AI refusal, followed shortly by a compliant AI response to a rephrased request within the same conversation session.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.
Detects anomalous data ingestion events in a Retrieval-Augmented Generation (RAG) pipeline. The rule flags 'Create' operations performed outside of scheduled batch ingestion jobs by identities that have not previously ingested data into the index. It further filters for ingested content that does not match any known document hashes within the existing corpus, identifying potential unauthorized or malicious document injection.
This rule detects high volumes of near-identical, incrementally-varied queries directed at an AI model API. This behavior is indicative of a black-box adversarial optimization loop, where an attacker attempts to infer model vulnerabilities, bypass safety filters, or perform model evasion by observing how slight input variations change model responses.