Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
2,832 detections
Filters
Last updated
All Time
Detection languages
1,375
495
363
281
175
Contributors
848
453
393
252
188
Categories
20,443
11,831
5,875
5,151
4,864
2,832
Platforms
1,449
916
693
670
264
Products / Services
994
701
613
469
216
MITRE Techniques
1,240
912
484
482
388
CVEs
6
5
2
1
1
IDS Classtypes
15
10
6
5
3
IDS Protocols
39
Detects anomalous AWS IAM and Secrets Manager API activity indicative of credential abuse following suspected exposure. The rule identifies suspicious patterns such as unauthorized enumeration of access keys, secret retrieval, and creation of new keys by non-service principals, reflecting activity associated with ShinyHunters-style operations where compromised credentials are abused.
Detects adversarial AI cost harvesting (AML.T0034.000) by identifying anomalous volumes of low-complexity queries against an AI inference API that result in a significant, unexplained spike in billed costs. This behavior mimics denial-of-service tactics aimed at exhausting compute capacity and financial resources.
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.
This rule detects anomalous behavior in AI agent tool invocations, specifically monitoring for high volumes of tool calls, high costs associated with token consumption, or suspicious tool interactions following a flagged indirect prompt injection. This behavior is indicative of potential resource exhaustion (denial of service) or abusive agent behavior as defined by the MITRE ATLAS framework for AI security.
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 anomalous access to AWS Bedrock Knowledge Bases, identifying a pattern of initial reconnaissance (enumerating bases and data sources) followed by high-volume querying (Retrieve/RetrieveAndGenerate). This sequence is indicative of an attacker attempting to map out backend storage configurations and subsequently exfiltrate indexed document content.
Detects unauthorized modification or deletion of AWS Bedrock guardrails. Adversaries may delete or degrade these guardrails (lowering sensitivity to 'NONE' or 'LOW') to bypass content filters, PII redaction, and safety controls, enabling toxic output generation or data exfiltration via LLM applications.
Detects the use of long-lived AWS IAM access keys (AKIA) to perform critical AWS Bedrock operations such as enabling foundation model access (PutUseCaseForModelAccess, CreateFoundationModelAgreement). Such sensitive configurations are typically expected to be performed via the AWS Console using federated credentials; the use of long-lived keys for these specific actions is a strong indicator of potential compromise and unauthorized resource enablement.
Detects unauthorized modifications to AWS Bedrock Agent configurations, specifically identifying high-risk activities such as rewriting system prompts for potential prompt injection or attaching cross-account Lambda functions. The rule monitors for UpdateAgent, CreateAgentActionGroup, UpdateAgentActionGroup, and PrepareAgent events.
Detects a suspicious sequence of identity-related operations performed by a single user within a 2-hour window. The chain consists of a password reset, MFA method modification, new device registration, and a subsequent privilege escalation event (such as role assignment or application consent). This pattern is consistent with helpdesk-vishing campaigns often used by threat actors like Scattered Spider to perform full account takeovers.
Detects anomalous GitHub audit log activity indicative of a potential supply-chain compromise. The rule identifies workflows or automated entities (Bot/App) that access sensitive Action secrets followed by the publishing of packages to public registries like PyPI or npm. This activity can represent the precursor steps of token-theft-driven malicious package injection.
This rule detects the creation of a new Azure AD (Entra ID) application with a name containing AI-related keywords (e.g., 'ai', 'copilot', 'gpt') that is subsequently granted high-privilege Microsoft Graph API permissions within a 24-hour window. This behavior is often associated with attackers setting up seemingly benign-looking applications to conduct OAuth-based persistent access or data exfiltration.
Detects a suspicious pattern where a user authenticates via device code and, within a short timeframe (two hours), registers a new device to their account. This behavior is indicative of the 'ARToken' persistence pattern, where an adversary uses a stolen session to bind a rogue device to the victim's account, potentially to mint a Primary Refresh Token (PRT) for persistent access.
Detects the 'auto-exfil' pattern associated with the TeamFiltration tool. The rule identifies a sequence of SharePoint Online file access immediately followed by a Microsoft Graph token request from the SharePoint Online Web Client Extensibility app. To reduce noise, it mandates a very short time delta between events (<= 60 seconds) combined with either the use of an IP address/ASN not previously seen for the account in the last 14 days, or a preceding failed sign-in attempt for the account.
Detects anomalous, high-volume reconnaissance activity against multiple Microsoft Graph API endpoints (users, groups, roles, authentication methods, service principals, sites, and drives) by a single identity within a one-hour window. This behavior is indicative of post-compromise enumeration, often associated with threats like UNC6671/Storm-3121.
Detects instances where a user successfully authenticates via the OAuth Device Code flow using an unmanaged or non-compliant device, followed closely by high-privileged Azure/Graph API administrative actions within a short timeframe. This behavior is indicative of a device-code phishing attack aimed at bypassing MFA controls, including FIDO2/passkeys, by compromising an active session.
Detects instances where a user successfully authenticates via the OAuth Device Code flow using an unmanaged or non-compliant device, followed closely by high-privileged Azure/Graph API administrative actions within a short timeframe. This behavior is indicative of a device-code phishing attack aimed at bypassing MFA controls, including FIDO2/passkeys, by compromising an active session.
This rule detects a multi-stage Business Email Compromise (BEC) attack involving executive impersonation via look-alike domains, followed by the creation of suspicious inbox rules (forwarding/deletion) by the victim, and concurrent external communication (Teams/Zoom meeting invites) from a newly created account. It correlates these activities to identify sophisticated payment fraud campaigns.
Detects attempts to tamper with, disable, or delete audit logging configurations in AWS CloudTrail and Azure Monitor. Monitoring these activities is critical to ensure that audit logs remain intact for security investigations.
Detects anomalous, high-frequency, multi-channel probing of AI agent input surfaces. The rule identifies external actors sending instruction-injection style keywords across different communication channels (e.g., email, Teams, calendar) within a short window, which may indicate an attempt to identify and manipulate an organization's autonomous agents. Covers T1595.002
Detects activity from known Storm-3168 (JADEPUFFER) campaign IP addresses across Azure control-plane, sign-in, and App Service logs. The rule monitors for malicious indicators in Azure Activity logs, Azure Sign-in logs, and Service Principal sign-ins, as well as specific sensitive URI paths (e.g., shells, admin login paths) within App Service HTTP logs.
Page 16 of 135