Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
2,982 detections
Filters
Last updated
All Time
Detection languages
1,486
513
403
239
167
Contributors
986
658
304
198
109
Categories
916
822
673
590
452
Platforms
39,934
7,265
6,535
3,865
3,645
2,982
Products / Services
1,229
703
443
309
200
MITRE Techniques
1,053
824
459
384
348
CVEs
17
16
16
10
10
IDS Classtypes
24
7
7
4
3
IDS Protocols
49
1
This rule detects internal users submitting potentially sensitive data to public generative AI applications (e.g., ChatGPT, Claude, Gemini) where Microsoft Purview DLP or a comparable inspection service has triggered a policy match. It correlates cloud application outbound signals with content inspection logs to identify high-risk exfiltration events while excluding sanctioned enterprise-licensed AI instances.
This rule detects potentially unauthorized device registration or Intune enrollment activity by monitoring Azure Sign-in logs. It specifically flags accounts that exhibit rapid, multiple device registration attempts (>= 2 in 5 minutes) combined with suspicious indicators such as scripted User Agents ('python-requests'), the presence of Primary Refresh Tokens (PRT), specific device naming patterns, or concurrent Intune enrollment activity. These behaviors are common indicators of automated adversary activity aiming to bypass conditional access policies by enrolling illicit devices.
Detects login attempts to source code repositories and developer tooling (GitHub, GitLab, Bitbucket, Azure DevOps, Slack, Visual Studio Code) originating from unmanaged or non-compliant devices, excluding traffic originating from known corporate network ranges.
Flags individual authentication events against services WaterPlum/North Korean IT worker operations are known to touch (laptop-farm remote-access tooling, freelance/crowdsourcing platforms, job-application and recruitment portals). This rule matches on a SINGLE qualifying event and does not itself prove IP reuse across contexts -- this Sigma dialect has no native cross-event correlation construct, so confirming that the SAME source IP appears across two or more of these categories (the actual WaterPlum indicator reported by investigators) requires a downstream correlation search, SOAR playbook, or UEBA rule that groups these flagged events by SourceIp over a rolling window.
Detects sign-in activity using the OAuth 2.0 device code flow, followed by subsequent authentication attempts from the same account originating from anomalous, previously unseen geographic locations or using unfamiliar applications. This pattern is indicative of a device code phishing attack where an adversary lures a user into authorizing a malicious device or application and subsequently hijacks the resulting session to access resources from a different location.
Detects repeated MFA authentication failures for specific Microsoft applications (like OfficeHome, My Apps, My Sign-Ins) from the same user and source IP, typically indicative of an MFA fatigue (prompt bombing) attack. The rule identifies multiple authentication failures or interruptions (ErrorCode 50074 or 50140) from unmanaged or unregistered devices, which suggests an adversary attempting to force a user to approve an MFA request.
Detects workflow or repository content referencing the compromised actions-cool/issues-helper or actions-cool/maintain-one-comment GitHub Actions, the known malicious commit, or the Bun-based execution of an obfuscated index.js payload
This rule detects scenarios where a device registration event occurs within 15 minutes of a device-code authentication sign-in for the same user. This pattern is indicative of potential account manipulation where an adversary registers their own device to an existing account to maintain persistent access or bypass multi-factor authentication (MFA).
This rule detects potential OAuth token theft resulting from device code phishing (e.g., N0va phishkit). It monitors for a successful OAuth device-code authentication followed within 5 minutes by a secondary successful sign-in for the same user from a different IP address, different browser client, or different user agent, which is a common indicator of an adversary-in-the-middle (AiTM) session hijacking attempt.
Detects anomalous activity associated with the OAuth 2.0 device code grant flow in Azure AD. By aggregating authentication requests within 15-minute windows, the rule identifies potential device code phishing or abuse, specifically flagging users exhibiting high-frequency polling (5 or more requests) or attempts from multiple distinct source IP addresses for a single session.
Detects instances where browser extensions, potentially AI-assisted (e.g., Claude, Gemini), trigger automated mail actions (send, forward, summarize) to non-standard, external domains. This behavior may indicate an unauthorized agent attempting to exfiltrate or manipulate email content.
Detects anomalous, high-volume file access or download activity in Microsoft SharePoint Online or OneDrive for Business originating from a 'python-httpx' user agent. This pattern often indicates automated data collection or exfiltration attempts using programmatic tools.
This rule detects a user account performing its first-ever GitHub OAuth or repository access within a short timeframe (30 minutes) after authenticating to a designated low-trust OpenAI community forum. This pattern is designed to identify potential adversary activity where a compromised or newly created account is used to pivot from a low-trust environment to corporate repository access.
Detects the granting of OAuth application permissions, specifically targeting applications with names containing terms like 'ai', 'copilot', or 'chatbot'. The rule looks for high-privilege scopes such as Mail, Files, or Directory access, which are common indicators of consent phishing or the use of malicious AI-branded applications to gain persistent access to sensitive data in the M365 environment.
Detects anomalous, high-volume file access or preview activity in OneDrive or SharePoint from non-browser user agents. This pattern is indicative of potential data exfiltration where adversaries use scripted clients or APIs to mass-read or preview files, bypassing traditional download-based alerts.
Detects anomalous activity associated with AI and LLM service providers by identifying accounts that are either interacting with these services for the first time or exhibiting a significant spike in API call volume compared to their 14-day historical baseline. This pattern may indicate unauthorized use of enterprise AI resources, potential persistence mechanisms, or exfiltration activities abusing legitimate LLM orchestration APIs.
Detects adversarial conversational reconnaissance patterns where a user queries an AI agent to enumerate its capabilities, such as available plugins, functions, or parameter schemas. The disclosure of this internal information by the agent facilitates the construction of targeted prompt-injection payloads, a known precursor to exploitation of AI agent vulnerabilities.
Detects potential post-authentication persistent actions (e.g., OAuth app grants, MFA changes, mailbox rule updates) performed shortly after a successful Google Workspace login, specifically flagging actions originating from IP addresses different from the login session to identify session hijacking or token replay (AiTM) activity. Includes suppression for known corporate IP ranges, trusted OAuth application consents, and recurring user behavior.
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 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 instances where a single user account exhibits an anomalously high frequency of requests to public generative-AI platforms within a one-hour window, potentially indicating automated reconnaissance or data staging activities using AI tools.
Page 43 of 142