Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
3,750 detections
Filters
Last updated
All Time
Detection languages
2,544
328
307
253
149
Contributors
953
720
641
297
173
Categories
1,245
994
880
773
476
Platforms
2,850
1,449
1,229
356
73
Products / Services
10,683
9,759
6,695
4,421
3,750
MITRE Techniques
1,096
989
950
912
820
CVEs
17
6
2
1
1
IDS Classtypes
9
4
3
2
2
IDS Protocols
21
1
Detects when Microsoft Cloud App Security (MCAS) reports a risky sign-in attempt due to an impossible travel scenario. This indicates a potential compromise of user credentials where an attacker logs in from a geographically distant location in a short timeframe, making physical travel impossible.
Detects internal phishing emails with subject lines mimicking IT service notifications, such as from ServiceNow. This is a tactic used by the Jingle Thief group (also tracked as Atlas Lion and STORM-0539) for lateral movement within a compromised organization.
The following analytic detects the use of default or publicly known named pipes associated with Cobalt Strike. It leverages Sysmon EventID 17 and 18 to identify specific named pipes commonly used by Cobalt Strike's Artifact Kit and Malleable C2 Profiles. This activity is significant because Cobalt Strike is a popular tool for adversaries to conduct post-exploitation tasks, and identifying its named pipes can reveal potential malicious activity. If confirmed malicious, this could indicate an active Cobalt Strike beacon, leading to unauthorized access, data exfiltration, or further lateral movement within the network.
Detects when a user successfully logs on to sanctioned applications from an IP address identified as risky by Microsoft Cloud App Security. This indicates a potential compromise of the user's account or an attempt to access resources from an unusual and potentially malicious location.
This analytic detects suspicious data transfer activities involving the use of the `split` syscall with the `-b` option, which specifies the size of each output file. This technique can be used by attackers to evade detection by breaking large files into smaller parts, bypassing size-based security controls, and facilitating covert exfiltration of sensitive data. By monitoring for unusual or unauthorized use of the `split` syscall with a specified block size, this analytic helps identify potential data exfiltration attempts.
KQL Query
This analytic identifies the bulk creation of ransomware notes (e.g., .txt, .html, .hta files) on an infected machine. It leverages Sysmon EventCode 11 to detect multiple instances of these file types being created within a short time frame (10 seconds). A detection is triggered if 15 or more unique readme paths are observed. This activity is a strong indicator of an active ransomware attack, where the attacker is notifying the victim of the encryption. If confirmed malicious, this behavior could lead to widespread data encryption, rendering critical files inaccessible and potentially causing significant operational disruption.
This analytic identifies the modification of the SourceAnchor (ImmutableId) attribute for an Azure Active Directory user. This detection leverages Azure AD audit logs, specifically monitoring the "Update user" operation and changes to the SourceAnchor attribute. This activity is significant as it is a step in setting up an Azure AD identity federation backdoor, allowing an adversary to establish persistence. If confirmed malicious, the attacker could impersonate any user, bypassing password and MFA requirements, leading to unauthorized access and potential data breaches.
The following analytic detects suspicious attempts to find credentials stored in password stores, indicating a potential attacker's effort to access sensitive login information. Password stores are critical repositories that contain valuable credentials, and unauthorized access to them can lead to significant security breaches. By monitoring for unusual or unauthorized activities related to password store access, this analytic helps identify potential credential theft attempts, allowing security teams to respond promptly and prevent unauthorized access to critical systems and data.
Detects M365 Copilot jailbreak attempts through prompt injection techniques including rule manipulation, system bypass commands, and AI impersonation requests that attempt to circumvent built-in safety controls. The detection searches exported eDiscovery prompt logs for jailbreak keywords like "pretend you are," "act as," "rules=", "ignore," "bypass," and "override" in the Subject_Title field, assigning severity scores based on the manipulation type (score of 4 for amoral impersonation or explicit rule injection, score of 3 for entity roleplay or bypass commands). Prompts with a jailbreak score of 2 or higher are flagged, prioritizing the most severe attempts to override AI safety mechanisms through direct instruction injection or unauthorized persona adoption.
Detects agentic AI jailbreak attempts that try to establish persistent control over M365 Copilot through rule injection, universal triggers, response automation, system overrides, and persona establishment techniques. The detection analyzes the PromptText field for keywords like "from now on," "always respond," "ignore previous," "new rule," "override," and role-playing commands (e.g., "act as," "you are now") that attempt to inject persistent instructions. The search computes risk by counting distinct jailbreak indicators per user session, flagging coordinated manipulation attempts.
Detects M365 Copilot users accessing from multiple geographic locations to identify potential account compromise, credential sharing, or impossible travel patterns. The detection aggregates M365 Copilot Graph API events per user, calculating distinct cities and countries accessed, unique IP addresses, and the observation timeframe to compute a locations-per-day metric that measures geographic mobility. Users accessing Copilot from more than one city (cities_count > 1) are flagged and sorted by country and city diversity, surfacing accounts exhibiting anomalous geographic patterns that suggest compromised credentials being used from distributed locations or simultaneous access from impossible travel distances.
Detects M365 Copilot access from non-compliant or unmanaged devices that violate corporate security policies, indicating potential shadow IT usage, BYOD policy violations, or compromised endpoint access. The detection filters M365 Copilot Graph API events where deviceDetail.isCompliant=false or deviceDetail.isManaged=false, then aggregates by user, operating system, and browser to calculate metrics including event counts, unique IPs and locations, and compliance/management status over time. Users accessing Copilot from non-compliant or unmanaged devices are flagged and sorted by activity volume and geographic spread, enabling security teams to identify unauthorized endpoints that may lack proper security controls, encryption, or MDM enrollment.
Detects M365 Copilot users exhibiting suspicious application usage patterns including multi-location access, abnormally high activity volumes, or access to multiple Copilot applications that may indicate account compromise or automated abuse. The detection aggregates M365 Copilot Graph API events per user, calculating metrics like distinct cities/countries accessed, unique IP addresses, number of different Copilot apps used, and average events per day over the observation period. Users are flagged when they access Copilot from multiple cities (cities_count > 1), generate excessive daily activity (events_per_day > 100), or use more than two different Copilot applications (app_count > 2), which are anomalous patterns suggesting credential compromise or bot-driven abuse.
This rule detects attempts to bypass M365 Copilot's safeguards and extract sensitive or comprehensive information using various social engineering techniques. It specifically looks for keywords in the 'Subject_Title' field of exported eDiscovery prompt logs that indicate requests for classified, proprietary, or bulk data, or attempts to bypass privacy controls. The rule categorizes these attempts by extraction type and assigns severity levels, filtering out low-severity cases to highlight the most critical risks of AI manipulation for data exfiltration.
This KQL is focus on gather statistics related to hackers activities over our tenants, such as the number of emails sent tagged as threats. Below, I’ve provided KQL queries to help identify the risky ASNs mentioned at the beggining.
You can adjust the filtering criteria by modifying the line: “| where SuspiciousPercentage > 95 and Total_different_IPs > 10” based on your specific requirements.
You can adjust the filtering criteria by modifying the line: “| where SuspiciousPercentage > 95 and Total_different_IPs > 10” based on your specific requirements.
Detects M365 Copilot users with failed authentication attempts, MFA failures, or multi-location access patterns indicating potential credential attacks or account compromise. The detection aggregates M365 Copilot Graph API authentication events per user, calculating metrics like distinct cities/countries accessed, unique IP addresses and browsers, failed login attempts (status containing "fail" or "error"), and MFA failures (error code 50074). Users are flagged when they access Copilot from multiple cities (cities_count > 1), experience any authentication failures (failed_attempts > 0), or encounter MFA errors (mfa_failures > 0), which are indicators of credential stuffing, brute force attacks, or compromised accounts attempting to bypass multi-factor authentication.
KQL Query
Detects M365 Copilot impersonation and roleplay jailbreak attempts where users try to manipulate the AI into adopting alternate personas, behaving as unrestricted entities, or impersonating malicious AI systems to bypass safety controls. The detection searches exported eDiscovery prompt logs for roleplay keywords like "pretend you are," "act as," "you are now," "amoral," and "roleplay as" in the Subject_Title field. Prompts are categorized into specific impersonation types (AI_Impersonation, Malicious_AI_Persona, Unrestricted_AI_Persona, etc.) to identify attempts to override the AI's safety guardrails through persona injection attacks.
The following analytic identifies the execution of RAR utilities to archive files on a system. It leverages data from Endpoint Detection and Response (EDR) agents, focusing on process names, GUIDs, and command-line arguments. This activity is significant as threat actors, including red-teamers and malware like DarkGate, use RAR archiving to compress and exfiltrate collected data from compromised hosts. If confirmed malicious, this behavior could lead to the unauthorized transfer of sensitive information to command and control servers, posing a severe risk to data confidentiality and integrity.
This rule detects suspicious sign-in attempts to Microsoft services (login.microsoftonline.com, graph.microsoft.com, outlook.office365.com) that occur within 30 minutes of a user clicking on a URL associated with these services. The sign-in is considered suspicious if it originates from a user agent commonly used by automated tools (e.g., axios, curl, python-requests, httpclient, Go-http-client) and is flagged with a 'atRisk' or 'high' risk level by Azure AD. This combination suggests a potential credential compromise following a phishing attempt, where an attacker might be using automated tools to leverage stolen credentials.
Page 142 of 179