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
9,773
6,704
4,421
3,749
1,084
MITRE Techniques
278
229
180
177
151
CVEs
17
15
13
13
8
IDS Classtypes
4
IDS Protocols
4
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 potential indirect prompt injection attacks against a Large Language Model (LLM) where untrusted context (e.g., RAG-retrieved documents, web content, or tool outputs) containing instruction-override or persona-switch keywords is ingested. This is followed by the LLM performing an anomalous privileged tool call within the same session, indicating a possible compromise of agent autonomy or unauthorized action execution.
Detects attempts to trigger a poisoned machine learning model through anomalous inference requests. The rule identifies inference API calls flagged as statistically anomalous by an upstream detector (e.g., unusual pixel patterns, token sequences, or steganographic payloads) that are correlated with a specific, spike-related output class across multiple distinct users.
Detects potential LLM jailbreak attempts by identifying a high frequency of guardrail-blocked responses followed by user prompts containing known jailbreak strategy fingerprints (such as roleplay, persona switching, or instruction overrides) within a 15-minute window.
Detects a sudden, anomalous spike in low-confidence actions or pending reviews originating from a single source within an AI agent action queue. This activity is indicative of a 'chaff' flooding attack, where an adversary submits a high volume of low-quality or irrelevant requests to overwhelm human-in-the-loop review systems, potentially causing a denial-of-service or masking malicious activity.
Detects potential adversarial machine learning attacks by monitoring for a concentration of input requests flagged with known adversarial patterns that result in statistically significant degradation of inference confidence scores for a specific ML model endpoint.
Detects potential Denial of Service (DoS) against AI/LLM inference endpoints. It monitors for high-frequency request spikes or abnormally large request payloads (oversized context windows) directed at common model endpoints that could lead to disproportionate compute resource consumption.
Detects anomalous usage patterns for AI-platform API keys consistent with account takeover, specifically identifying concurrent multi-geographic access and the exercise of elevated administrative, write, or model-registry scopes within a short timeframe.
Detects high-frequency, automated API requests to commercial Large Language Model (LLM) endpoints from a single source IP. The rule identifies suspicious activity characterized by a rapid rotation of multiple API tokens, often indicative of an intermediary proxy or middleware service attempting to bypass provider-specific usage limits, billing controls, or safety guardrails.
Detects instances where an AI agent invokes sensitive administrative or system tools (shell execution, file writing, network communication, or email sending) shortly after ingesting untrusted external content, in the absence of explicit, logged user instructions for those specific actions. This behavior is a primary indicator of indirect prompt injection attacks aimed at achieving command execution or C2-like behavior.
This rule detects potential prompt injection attacks against internal AI-powered assistants, agents, or LLM gateways. It monitors network HTTP requests for common patterns used to override system instructions, bypass safety filters, or force the model to adopt a privileged, unauthorized persona (e.g., administrator or debug mode). These techniques are often used in social engineering to manipulate the AI's output or security posture.
Detects a suspicious sequence of events where a user engagement (voice/video call) involving themes of urgency or helpdesk interaction is closely followed by a high-risk administrative action (e.g., password reset, MFA device enrollment, or financial transfer) by the same user, where the communication session context differs from standard activity, indicative of a deepfake-enabled vishing attack.
Detects high-frequency authentication attempts originating from user agents associated with automation frameworks (Selenium, Playwright, Puppeteer, etc.). This pattern is often indicative of automated credential stuffing, brute force attacks, or CAPTCHA/2FA bypass attempts orchestrated by headless browsers.
Detects malicious Android application installation or launch behavior consistent with the 'Corp MDM' implant. This rule specifically identifies instances where a package named 'com.corp.mdm' is installed or launched and rapidly requests a combination of high-risk permissions including SMS and telephony access (RECEIVE_SMS, READ_SMS, CALL_PHONE) alongside notification permissions, which is characteristic of the implant's initial setup phase.
Detects a suspicious sequence of events where a user engagement (voice/video call) involving themes of urgency or helpdesk interaction is closely followed by a high-risk administrative action (e.g., password reset, MFA device enrollment, or financial transfer) by the same user, where the communication session context differs from standard activity, indicative of a deepfake-enabled vishing attack.
Detects high-frequency authentication attempts originating from user agents associated with automation frameworks (Selenium, Playwright, Puppeteer, etc.). This pattern is often indicative of automated credential stuffing, brute force attacks, or CAPTCHA/2FA bypass attempts orchestrated by headless browsers.
This rule monitors Salesforce lead creation events for potential prompt-injection payloads. It identifies submissions containing common AI manipulation instructions, HTML tags, or suspicious outbound URLs, which could be used to hijack internal user sessions or execute malicious content within the Salesforce environment.
This rule monitors Salesforce lead creation events for potential prompt-injection payloads. It identifies submissions containing common AI manipulation instructions, HTML tags, or suspicious outbound URLs, which could be used to hijack internal user sessions or execute malicious content within the Salesforce environment.
This rule monitors Salesforce lead creation events for potential prompt-injection payloads. It identifies submissions containing common AI manipulation instructions, HTML tags, or suspicious outbound URLs, which could be used to hijack internal user sessions or execute malicious content within the Salesforce environment.
This rule detects Salesforce Agentforce or similar AI-driven chat responses that contain external URLs with non-standard Top-Level Domains (TLDs). It identifies cases where the organization's redactor failed to strip a URL due to unexpected TLDs or unusual trailing termination characters (e.g., brackets) that are still rendered as clickable links in the interface. Such links may facilitate phishing, credential harvesting, or redirection to malicious sites.
Page 18 of 52