Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,733 detections
Filters
Last updated
All Time
Detection languages
909
387
164
127
68
Contributors
479
417
201
177
61
Categories
20,443
11,831
5,875
5,151
4,864
1,733
Platforms
822
436
329
274
206
Products / Services
498
378
118
96
79
MITRE Techniques
433
327
225
196
196
CVEs
12
12
10
10
6
IDS Classtypes
1
1
IDS Protocols
2
Detects adversarial prompts designed to induce an Large Language Model (LLM) to disclose sensitive information, such as PII, secrets, or internal training data. This rule monitors the input prompt for extraction-based patterns and evaluates the model's response for indicators of a sensitive data leak (DLP hit).
Detects attempts to perform a Denial of Service (DoS) attack against an AI-enabled system by flooding the inference API with high volumes of requests or overly large inputs (high token counts). The rule identifies potential resource exhaustion by correlating high request rates or large input sizes with system-wide impacts, such as increased inference queue depth or excessive GPU utilization, which could impede service availability for other users.
This rule detects attempts to bypass LLM safety guardrails using adversarial prompting techniques (e.g., 'DAN' persona, hypothetical scenarios) or by identifying patterns of repeated moderation flags and refusals. It specifically looks for user prompts containing known jailbreak templates or an excessive frequency of safety-related trigger events, while excluding authorized red-team and safety evaluation accounts.
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 unauthorized or anomalous modifications to machine learning model checkpoints. The rule triggers on significant deviations in weight delta percentages outside of scheduled fine-tuning windows, or when a model checkpoint hash is altered by users or service accounts without corresponding CI/CD pipeline approval or documented code review.
Detects suspicious AI agent behavior by identifying two primary scenarios: 1) The invocation of sensitive tools (e.g., code execution, file deletion) within a short timeframe following a flagged prompt injection or jailbreak attempt in the same conversation; or 2) The invocation of a tool by an AI agent that is not explicitly defined in its allowed tool list for the given workflow.
Detects activity associated with the SharePoint ToolShell exploitation chain, specifically monitoring the w3wp.exe process for spawning suspicious child processes like cmd.exe or powershell.exe, and the creation of unexpected .aspx files within the SharePoint LAYOUTS directory.
Detects outbound network requests from non-standard Android application processes to Google's generative-AI API endpoints (e.g., Gemini). This behavior is indicative of malware, such as RatHat, attempting to exfiltrate accessibility information or UI trees to an LLM to facilitate automated screen interaction and navigation.
Detects outbound network requests from non-standard Android application processes to Google's generative-AI API endpoints (e.g., Gemini). This behavior is indicative of malware, such as RatHat, attempting to exfiltrate accessibility information or UI trees to an LLM to facilitate automated screen interaction and navigation.
Detects anomalous, high-volume automated API querying or bulk data exports within SaaS environments that target keywords associated with sensitive information, such as passwords, tokens, or configuration secrets. This pattern is characteristic of automated reconnaissance and unauthorized data harvesting, often observed in campaigns like those attributed to UNC6395.
Detects suspicious HTTP requests directed at SharePoint applications containing a malformed Register directive. The rule identifies attempts to leverage the 'publishingribbon' Tagprefix combined with unauthorized namespaces (e.g., System.Xaml, System.Windows.Data) to bypass the EditingPageParser.InitializeRegisterTable safelist check, a common technique for SharePoint remote code execution.
Detects anomalous OAuth application consent activity by identifying a surge in unique users granting permissions to a single OAuth application within a short timeframe. This behavior is indicative of consent phishing campaigns where attackers lure users to grant application access to malicious apps, effectively stealing application access tokens and gaining persistent access to resources without requiring the victim's password.
Detects instances where a non-administrative user grants OAuth application permissions that include high-risk delegated scopes. This activity is a primary indicator of illicit consent grant phishing attacks, where an adversary lures users into authorizing malicious applications to access sensitive data (such as emails or files) without administrator approval.
Detects OAuth application consent events where an application is granted a combination of 'offline_access' (allowing refresh token issuance) and email-read permissions. This pattern is commonly used by adversaries to establish persistent, long-term access to user mailboxes for data exfiltration or Business Email Compromise (BEC) activities, surviving password changes until the consent is explicitly revoked.
Detects instances where a user has consented to an OAuth application that requests high-privilege Microsoft Graph scopes (Directory.Read.All, User.Read.All, Mail.Read). This pattern is indicative of consent phishing attacks where adversaries trick users into granting applications excessive permissions, bypassing administrative consent workflows when tenant configurations allow user-level consent.
Detects mass-phishing consent campaigns by monitoring non-admin users consenting to multi-tenant or unverified-publisher applications requesting non-basic-profile OAuth scopes. The rule aggregates multiple consent events for the same application over a 1-hour window to identify potentially malicious consent-phishing activity.
Detects OAuth2 permission consent events where a user grants a third-party or newly registered application access to high-risk scopes (such as Mail, Files, or Directory). This technique is commonly used to establish persistence and bypass MFA by gaining access to user accounts via malicious or compromised OAuth applications.
Detects OAuth2 permission consent events where a user grants a third-party or newly registered application access to high-risk scopes (such as Mail, Files, or Directory). This technique is commonly used to establish persistence and bypass MFA by gaining access to user accounts via malicious or compromised OAuth applications.
Detects accounts showing a spike in requests to public generative-AI services compared to their historical baseline, combined with anomalous indicators like multiple source IPs, regions, or off-hours activity.
Detects browser extension updates that request a combination of high-risk permissions capable of account takeover (e.g., cookies, webRequest, identity) for extensions previously classified as low-risk. This pattern often indicates a supply chain compromise where a benign extension is acquired and subsequently updated with malicious capabilities.
Detects browser extension updates that request a combination of high-risk permissions capable of account takeover (e.g., cookies, webRequest, identity) for extensions previously classified as low-risk. This pattern often indicates a supply chain compromise where a benign extension is acquired and subsequently updated with malicious capabilities.
Page 7 of 83