EXPLORE DETECTIONS
M365 Copilot Impersonation Jailbreak Attack
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.
M365 Copilot Information Extraction Jailbreak Attack
Detects M365 Copilot information extraction jailbreak attacks that attempt to obtain sensitive, classified, or comprehensive data through various social engineering techniques including fictional entity impersonation, bulk data requests, and privacy bypass attempts. The detection searches exported eDiscovery prompt logs for extraction keywords like "transcendent," "tell me everything," "confidential," "dump," "extract," "reveal," and "bypass" in the Subject_Title field, categorizing each attempt by extraction type and assigning severity levels (CRITICAL for classified/proprietary data, HIGH for bulk extraction or privacy bypass). Prompts are further analyzed for compound risk patterns such as "Confidential+Extraction" or "Bulk_Request+Bypass," filtering out low-severity cases to surface the most dangerous attempts to exfiltrate sensitive organizational information through AI manipulation.
M365 Copilot Jailbreak Attempts
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.
M365 Copilot Non Compliant Devices Accessing M365 Copilot
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.
M365 Copilot Session Origin Anomalies
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.
MacOS - Re-opened Applications
The following analytic identifies processes referencing plist files that determine which applications are re-opened when a user reboots their MacOS machine. It leverages data from Endpoint Detection and Response (EDR) agents, focusing on process names and parent processes related to "com.apple.loginwindow." This activity is significant because it can indicate attempts to persist across reboots, a common tactic used by attackers to maintain access. If confirmed malicious, this could allow an attacker to execute code or maintain persistence on the affected system, potentially leading to further compromise.
MacOS Account Created
The following analytic detects the creation of a new local user account on a MacOS system. It leverages osquery logs to identify this activity. Monitoring the creation of local accounts is crucial for a SOC as it can indicate unauthorized access or lateral movement within the network. If confirmed malicious, this activity could allow an attacker to establish persistence, escalate privileges, or gain unauthorized access to sensitive systems and data.
MacOS AMOS Stealer - Virtual Machine Check Activity
The following analytic detects AMOS Stealer VM check activity on macOS. It leverages osquery to monitor process events and identifies the execution of the "osascript" command along with specific commandline strings. This activity is significant as AMOS stealer was seen using this pattern in order to check if the host is a Virtual Machine or not. If confirmed malicious, this behavior indicate that the host is already infected by the AMOS stealer, which could allow attackers to execute arbitrary code, escalate privileges, steal information, or persist within the environment, posing a significant security risk.
MacOS Data Chunking
The following analytic detects suspicious data chunking activities that involve the use of split or dd, potentially indicating an attempt to evade detection by breaking large files into smaller parts. Attackers may use this technique to bypass size-based security controls, facilitating the covert exfiltration of sensitive data. By monitoring for unusual or unauthorized use of these commands, this analytic helps identify potential data exfiltration attempts, allowing security teams to intervene and prevent the unauthorized transfer of critical information from the network.
MacOS Gatekeeper Bypass
Detects known MacOS security bypass techniques that may be used to enable malicious code execution. Specifically monitors for attempts to remove the com.apple.quarantine attribute using xattr, or to disable Gatekeeper protections via spctl --master-disable, both of which can allow untrusted or malicious applications to execute without standard system safeguards.
MacOS Hidden Files and Directories
The following analytic detects suspicious creation of hidden files and directories, which may indicate an attacker's attempt to conceal malicious activities or unauthorized data. Hidden files and directories are often used to evade detection by security tools and administrators, providing a stealthy means for storing malware, logs, or sensitive information. By monitoring for unusual or unauthorized creation of hidden files and directories, this analytic helps identify potential attempts to hide or unauthorized creation of hidden files and directories, and helps identify potential attempts to hide malicious operations, enabling security teams to uncover and address hidden threats effectively.
MacOS Kextload Usage
Detects execution of the kextload command on macOS systems. The kextload utility is used to manually load kernel extensions (KEXTs) into the macOS kernel, which can introduce privileged code at the kernel level. While legitimate for driver installation and system administration, misuse may indicate attempts to install unauthorized, malicious, or persistence-enabling kernel extensions.
MacOS Keychains Dumped
Detects command-line attempts to access or dump macOS Keychain data using native utilities or direct file access. This includes credential dumping via the `security` utility (e.g. `dump-keychain -d`), bulk certificate export using `security find-certificate`, and direct file copying of Keychain database files using utilities such as `cat`. Keychain files are located in `~/Library/Keychains/`, `/Library/Keychains/`, and `/Network/Library/Keychains/`. This technique is commonly associated with post-exploitation credential harvesting, where an attacker with local access seeks to escalate privileges or move laterally by obtaining stored credentials for applications, Wi-Fi networks, system services, and certificates.
MacOS List Firewall Rules
This analytic detects attempts to enumerate or verify the configuration of the macOS application firewall. Specifically, it monitors executions of `defaults read /Library/Preferences/com.apple.alf` and `/usr/libexec/ApplicationFirewall/socketfilterfw --getglobalstate`. These commands provide insight into firewall status, allowed applications, and explicit authorization rules. While they are legitimate administrative operations, adversaries may leverage them to identify potential attack surfaces, determine whether the firewall is active, or enumerate allowed network flows. Monitoring for these commands, particularly when executed by non-administrative users or at unusual times, can provide early indication of reconnaissance activity on macOS endpoints
MacOS Log Removal
Detects the deletion or modification of logs on MacOS systems by identifying execution of the rm command with command-line arguments referencing system.log or audit-related paths. Adversaries may remove or alter log files to cover their tracks and hinder detection and forensic analysis. This behavior commonly occurs during post-exploitation cleanup.
MacOS LoginHook Persistence
Identifies attempts to configure a macOS LoginHook via the defaults utility. LoginHooks enable automatic execution of a script or program upon user login and have historically been abused for persistence. Creation or modification of this setting may indicate an attempt to establish startup execution outside standard LaunchAgent mechanisms.
MacOS LOLbin
The following analytic detects multiple executions of Living off the Land (LOLbin) binaries on macOS within a short period. It leverages osquery to monitor process events and identifies commands such as "find", "crontab", "screencapture", "openssl", "curl", "wget", "killall", and "funzip". This activity is significant as LOLbins are often used by attackers to perform malicious actions while evading detection. If confirmed malicious, this behavior could allow attackers to execute arbitrary code, escalate privileges, or persist within the environment, posing a significant security risk.
MacOS Network Share Discovery
Identifies execution of network share enumeration commands (smbutil, showmount) that can be leveraged by adversaries to discover accessible SMB and NFS resources, supporting internal reconnaissance and potential lateral movement.
MacOS plutil
The following analytic detects the usage of the `plutil` command to modify plist files on macOS systems. It leverages osquery to monitor process events, specifically looking for executions of `/usr/bin/plutil`. This activity is significant because adversaries can use `plutil` to alter plist files, potentially adding malicious binaries or command-line arguments that execute upon user logon or system startup. If confirmed malicious, this could allow attackers to achieve persistence, execute arbitrary code, or escalate privileges, posing a significant threat to the system's security.
Mailsniper Invoke functions
The following analytic detects the execution of known MailSniper PowerShell functions on a machine. It leverages PowerShell logs (EventCode 4104) to identify specific script block text associated with MailSniper activities. This behavior is significant as MailSniper is often used by attackers to harvest sensitive emails from compromised Exchange servers. If confirmed malicious, this activity could lead to unauthorized access to sensitive email data, credential theft, and further compromise of the email infrastructure.
Malicious InProcServer32 Modification
The following analytic detects a process modifying the registry with a known malicious CLSID under InProcServer32. It leverages data from Endpoint Detection and Response (EDR) agents, focusing on registry modifications within the HKLM or HKCU Software Classes CLSID paths. This activity is significant as it may indicate an attempt to load a malicious DLL, potentially leading to code execution. If confirmed malicious, this could allow an attacker to persist in the environment, execute arbitrary code, or escalate privileges, posing a severe threat to system integrity and security.
Malicious Powershell Executed As A Service
The following analytic identifies the execution of malicious PowerShell commands or payloads via the Windows SC.exe utility. It detects this activity by analyzing Windows System logs (EventCode 7045) and filtering for specific PowerShell-related patterns in the ImagePath field. This behavior is significant because it indicates potential abuse of the Windows Service Control Manager to run unauthorized or harmful scripts, which could lead to system compromise. If confirmed malicious, this activity could allow attackers to execute arbitrary code, escalate privileges, or maintain persistence within the environment.
Malicious PowerShell Process - Encoded Command
The following analytic detects the use of the EncodedCommand parameter in PowerShell processes. It leverages Endpoint Detection and Response (EDR) data to identify variations of the EncodedCommand parameter, including shortened forms and different command switch types. This activity can be significant because adversaries often use encoded commands to obfuscate malicious scripts, making detection harder. If confirmed malicious, this behavior could allow attackers to execute hidden code, potentially leading to unauthorized access, privilege escalation, or persistent threats within the environment. Review parallel events to determine legitimacy and tune based on known administrative scripts.
Malicious PowerShell Process - Execution Policy Bypass
The following analytic detects PowerShell processes initiated with parameters that bypass the local execution policy for scripts. It leverages data from Endpoint Detection and Response (EDR) agents, focusing on command-line executions containing specific flags like "-ex" or "bypass." This activity is significant because bypassing execution policies is a common tactic used by attackers to run malicious scripts undetected. If confirmed malicious, this could allow an attacker to execute arbitrary code, potentially leading to further system compromise, data exfiltration, or persistent access within the environment.