Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,907 detections
Filters
Last updated
All Time
Detection languages
719
486
238
182
142
Contributors
547
475
198
184
152
Categories
1,535
958
955
583
128
Platforms
1,890
19
10
7
5
Products / Services
10,665
9,773
6,705
4,421
3,749
1,907
MITRE Techniques
1,254
1,146
731
283
260
CVEs
7
3
2
2
2
IDS Classtypes
1
IDS Protocols
1
The following analytic detects the execution of the `Get-DomainComputer` commandlet using PowerShell Script Block Logging (EventCode=4104). This commandlet is part of PowerView, a tool often used for enumerating domain computers within Windows environments. The detection leverages script block text analysis to identify this specific command. Monitoring this activity is crucial as it can indicate an adversary's attempt to gather information about domain computers, which is a common step in Active Directory reconnaissance. If confirmed malicious, this activity could lead to further network enumeration and potential lateral movement within the domain.
The analytic detects the execution of the `GetCurrent` method from the WindowsIdentity .NET class using PowerShell Script Block Logging (EventCode=4104). This method identifies the current Windows user. Adversaries and Red Teams may use this for situational awareness and Active Directory discovery on compromised endpoints to map user accounts, potentially leading to privilege escalation or lateral movement.
The following analytic detects the execution of the `Get-DomainGroup` cmdlet using PowerShell Script Block Logging (EventCode=4104). This cmdlet, part of the PowerView tool, is used to enumerate domain groups within a Windows domain. The detection leverages script block text to identify this specific command. Monitoring this activity is crucial as it may indicate an adversary or Red Team performing reconnaissance to gain situational awareness and map out Active Directory structures. If confirmed malicious, this activity could lead to further exploitation, including privilege escalation and lateral movement within the network.
The analytic detects the execution of the `Get-WmiObject` cmdlet with the `DS_Computer` class parameter via PowerShell Script Block Logging (EventCode=4104). This detection leverages script block text to identify queries targeting domain computers using WMI. Monitoring this activity is crucial as adversaries and Red Teams may use it for Active Directory Discovery and situational awareness. If confirmed malicious, this behavior could allow attackers to map out domain computers, facilitating further attacks such as lateral movement or privilege escalation.
This analytic detects suspicious PowerShell script execution that utilizes memory streams as a backing store, identified through EventCode 4104 (PowerShell Script Block Logging). It specifically looks for scripts that create new objects with 'IO.MemoryStream', a technique often employed to decompress and execute payloads directly in memory. This activity is significant as it indicates potential in-memory execution of malicious code, which can bypass traditional file-based detection mechanisms. If confirmed malicious, this technique could allow attackers to execute arbitrary code, maintain persistence, or escalate privileges without leaving traces on disk.
The following analytic detects the execution of the PowerView PowerShell cmdlet Get-NetUser, specifically querying for "samaccountname" and "pwdlastset" attributes. It leverages Event ID 4104 from PowerShell Script Block Logging to identify this activity. This behavior is significant as it may indicate an attempt to gather user account information from Active Directory, which is a common reconnaissance step in lateral movement or privilege escalation attacks. If confirmed malicious, this activity could allow an attacker to map out user accounts, potentially leading to further exploitation and unauthorized access within the network.
The following analytic detects the execution of the PowerView PowerShell cmdlet Get-NetUser with the UACFilter parameter set to NOT_ACCOUNTDISABLE, indicating an attempt to enumerate Active Directory user accounts that are not disabled. This detection leverages PowerShell Script Block Logging (EventCode 4104) to identify the specific script block text. Monitoring this activity is significant as it may indicate reconnaissance efforts by an attacker to identify active user accounts for further exploitation. If confirmed malicious, this activity could lead to unauthorized access, privilege escalation, or lateral movement within the network.
This analytic detects the execution of the `Get-LocalUser` PowerShell commandlet using PowerShell Script Block Logging (EventCode=4104). This commandlet lists all local users on a system. Monitoring this behavior is significant as adversaries and Red Teams may use it to enumerate local users for situational awareness and Active Directory discovery. If confirmed malicious, this activity could lead to further reconnaissance, enabling attackers to identify potential targets for privilege escalation or lateral movement.
This analytic detects the execution of the `Get-WmiObject` cmdlet with the `Win32_UserAccount` parameter via PowerShell Script Block Logging (EventCode=4104). This method identifies when a list of all local users is being enumerated. This activity is significant as it may indicate an adversary or Red Team operation attempting to gather user information for situational awareness and Active Directory discovery. If confirmed malicious, this could lead to further reconnaissance, privilege escalation, or lateral movement within the network.
This analytic detects the execution of a PowerShell script that modifies the 'TrustedHosts' configuration via EventCode 4104. It leverages PowerShell Script Block Logging to identify commands targeting WSMan settings, specifically those altering or concatenating trusted hosts. This activity is significant as it can indicate attempts to manipulate remote connection settings, potentially allowing unauthorized remote access. If confirmed malicious, this could enable attackers to establish persistent remote connections, bypass security protocols, and gain unauthorized access to sensitive systems and data.
This analytic detects PowerShell commands that initiate a process on a remote endpoint using the Distributed Component Object Model (DCOM) protocol. It specifically looks for the use of "ShellExecute" or "ExecuteShellCommand" within PowerShell Script Block Logging (EventCode=4104). This activity is significant as it may indicate lateral movement or remote code execution attempts by adversaries, potentially leading to further compromise and persistence within the network.
The following analytic detects the execution of the PowerShell cmdlet `get-localgroup` using PowerShell Script Block Logging (EventCode=4104). This method captures the full command sent to PowerShell, providing detailed visibility into script execution. Monitoring this activity is significant as it can indicate an attempt to enumerate local groups, which may be a precursor to privilege escalation or lateral movement. If confirmed malicious, an attacker could gain insights into group memberships, potentially leading to unauthorized access or privilege abuse. Review parallel processes and the entire script block for comprehensive analysis.
The following analytic identifies the use of PowerShell's Start-Service or Stop-Service cmdlets on an endpoint. It leverages PowerShell Script Block Logging to detect these commands. This activity is significant because attackers can manipulate services to disable or stop critical functions, causing system instability or disrupting business operations. If confirmed malicious, this behavior could allow attackers to disable security services, evade detection, or disrupt essential services, leading to potential system downtime and compromised security.
This analytic detects the execution of PowerShell commands that use the `Invoke-Command` cmdlet to start a process on a remote endpoint via the WinRM protocol. It leverages PowerShell Script Block Logging (EventCode=4104) to identify such activities. This behavior is significant as it may indicate lateral movement or remote code execution attempts by adversaries. If confirmed malicious, this activity could allow attackers to execute arbitrary code on remote systems, potentially leading to further compromise and persistence within the network.
This analytic detects the use of PowerShell environment variables to identify the current logged-in user by leveraging PowerShell Script Block Logging (EventCode=4104). Specifically, it monitors script blocks containing `$env:UserName` or `[System.Environment]::UserName`. This activity is significant because adversaries and Red Teams often use it for situational awareness and Active Directory discovery on compromised endpoints. If confirmed malicious, this activity could allow attackers to gain insights into user context, aiding in further exploitation and lateral movement within the network.
Baseline Comparison.
Author: miflower.
The purpose of this query is to perform a comparison between "known good" machines and suspected bad machines.
The original concept for this query was born due to reapplying the same 'whitelist' filters over and over.
It brings deltas between a baseline and another machine quickly to the analyst's view.
This query supports multiple suspected bad machines and multiple "known good" machines.
It also supports providing a timeframe for how far back in time to build a baseline as well as how far back in time to evaluate the suspected bad machines.
Each of the links provided by DeviceId/DeviceName will go to the most recent entry for whatever entity is listed.
Average results for the pre-defined settings below with a single good host and a single bad host on a 'huge' tenant (300k+ machines):.
Compute Time: ~10-20 seconds.
Result Set Size: ~500 rows.
The workflow is as follows:.
1. Establish Variables that are editable on a per-query basis.
2. Define functions for reuse.
3. Calculate DeviceIds for all machines in scope.
4. Derive deltas using the aforementioned functions.
5. Union together all results into a single view.
The following datasets are returned:.
1. Alerts on the suspected bad machines (ignores known good machines, because...they're alerts, additional data has the triggered file).
2. Connected Networks (from DeviceNetworkInfo table, additional data has full Connected Network details).
3. File Creations (disabled by default due to volume, enable at your own risk, additional data has initiating processes).
4. Image Loads (disabled by default due to volume, enable at your own risk, additional data has initiating processes).
5. Logon (derived from DeviceLogonEvents for the unique users logged on, additional data has logon types).
6. Network communication (grouped by 2nd level-domain, ie 'microsoft.com' in 'www.microsoft.com' and 'web.microsoft.com', additional data has the full list of URLs).
7. Process creation (additional da
Author: miflower.
The purpose of this query is to perform a comparison between "known good" machines and suspected bad machines.
The original concept for this query was born due to reapplying the same 'whitelist' filters over and over.
It brings deltas between a baseline and another machine quickly to the analyst's view.
This query supports multiple suspected bad machines and multiple "known good" machines.
It also supports providing a timeframe for how far back in time to build a baseline as well as how far back in time to evaluate the suspected bad machines.
Each of the links provided by DeviceId/DeviceName will go to the most recent entry for whatever entity is listed.
Average results for the pre-defined settings below with a single good host and a single bad host on a 'huge' tenant (300k+ machines):.
Compute Time: ~10-20 seconds.
Result Set Size: ~500 rows.
The workflow is as follows:.
1. Establish Variables that are editable on a per-query basis.
2. Define functions for reuse.
3. Calculate DeviceIds for all machines in scope.
4. Derive deltas using the aforementioned functions.
5. Union together all results into a single view.
The following datasets are returned:.
1. Alerts on the suspected bad machines (ignores known good machines, because...they're alerts, additional data has the triggered file).
2. Connected Networks (from DeviceNetworkInfo table, additional data has full Connected Network details).
3. File Creations (disabled by default due to volume, enable at your own risk, additional data has initiating processes).
4. Image Loads (disabled by default due to volume, enable at your own risk, additional data has initiating processes).
5. Logon (derived from DeviceLogonEvents for the unique users logged on, additional data has logon types).
6. Network communication (grouped by 2nd level-domain, ie 'microsoft.com' in 'www.microsoft.com' and 'web.microsoft.com', additional data has the full list of URLs).
7. Process creation (additional da
' This hunting query looks for process command line activity related to activity observed by Dev-0056.The command lines this query hunts for are used as part of the threat actor's post exploitation activity.
This query uses the Microsoft Sentinel Information Model - https://docs.microsoft.com/azure/sentinel/normalization'
This query uses the Microsoft Sentinel Information Model - https://docs.microsoft.com/azure/sentinel/normalization'
This query identifies common processes run by ransomware
malware to destroy volume shadow copies or clean free
space on a drive to prevent a file from being recovered
post-encryption. To reduce false positives, results are
filtered to only actions taken when the initiating
process was launched from a suspicious directory. If
you don't mind false positives, consider removing the
last where clause.
Special thanks to Captain for additional inputs
malware to destroy volume shadow copies or clean free
space on a drive to prevent a file from being recovered
post-encryption. To reduce false positives, results are
filtered to only actions taken when the initiating
process was launched from a suspicious directory. If
you don't mind false positives, consider removing the
last where clause.
Special thanks to Captain for additional inputs
This query was originally published in the threat analytics report, "Exchange Server zero-days exploited in the wild".
In early March 2021, Microsoft released patches for four different zero-day vulnerabilities affecting Microsoft Exchange Server. The vulnerabilities were being used in a coordinated attack. For more information on the vulnerabilities, visit the following links:
1. CVE-2021-26855
2. CVE-2021-26857
3. CVE-2021-26858
4. CVE-2021-27065
The following query detects downloads of powercat, an exploitation tool for PowerShell. Although associated with these zero-day attacks, powercat is a multi-purpose tool that is also used by other groups of attackers.
More queries related to this threat can be found under the See also section of this page.
Reference - https://msrc-blog.microsoft.com/2021/03/02/multiple-security-updates-released-for-exchange-server/
In early March 2021, Microsoft released patches for four different zero-day vulnerabilities affecting Microsoft Exchange Server. The vulnerabilities were being used in a coordinated attack. For more information on the vulnerabilities, visit the following links:
1. CVE-2021-26855
2. CVE-2021-26857
3. CVE-2021-26858
4. CVE-2021-27065
The following query detects downloads of powercat, an exploitation tool for PowerShell. Although associated with these zero-day attacks, powercat is a multi-purpose tool that is also used by other groups of attackers.
More queries related to this threat can be found under the See also section of this page.
Reference - https://msrc-blog.microsoft.com/2021/03/02/multiple-security-updates-released-for-exchange-server/
This query was originally published in the threat analytics report, Ryuk ransomware. There is also a related blog.
Ryuk is human-operated ransomware. Much like DoppelPaymer ransomware, Ryuk is spread manually, often on networks that are already infected with Trickbot.
During the earliest stages of a Ryuk infection, an operator downloads Cobalt Strike, a penetration testing kit that is also used by malicious actors. Cobalt Strike is used by Ryuk operators to explore the network before deploying the Ryuk payload. This malicious behavior is often obscured by Base64 encoding and other tricks.
The following query detects possible invocation of Cobalt Strike using Windows Management Instrumentation (WMI).
The See also section below lists links to other queries associated with Ryuk ransomware.
References:
https://www.microsoft.com/security/blog/2020/03/05/human-operated-ransomware-attacks-a-preventable-disaster/
https://www.microsoft.com/wdsi/threats/malware-encyclopedia-description?Name=Ransom:Win32/Ryuk&threatId=-2147232689
https://www.microsoft.com/security/blog/2020/03/05/human-operated-ransomware-attacks-a-preventable-disaster/
https://www.cobaltstrike.com/
https://docs.microsoft.com/windows/win32/wmisdk/wmi-start-page
Ryuk is human-operated ransomware. Much like DoppelPaymer ransomware, Ryuk is spread manually, often on networks that are already infected with Trickbot.
During the earliest stages of a Ryuk infection, an operator downloads Cobalt Strike, a penetration testing kit that is also used by malicious actors. Cobalt Strike is used by Ryuk operators to explore the network before deploying the Ryuk payload. This malicious behavior is often obscured by Base64 encoding and other tricks.
The following query detects possible invocation of Cobalt Strike using Windows Management Instrumentation (WMI).
The See also section below lists links to other queries associated with Ryuk ransomware.
References:
https://www.microsoft.com/security/blog/2020/03/05/human-operated-ransomware-attacks-a-preventable-disaster/
https://www.microsoft.com/wdsi/threats/malware-encyclopedia-description?Name=Ransom:Win32/Ryuk&threatId=-2147232689
https://www.microsoft.com/security/blog/2020/03/05/human-operated-ransomware-attacks-a-preventable-disaster/
https://www.cobaltstrike.com/
https://docs.microsoft.com/windows/win32/wmisdk/wmi-start-page
'This query looks for command lines that contain a public IP address. Attackers may use a hard coded IP for C2 or exfiltration.
This query can be filtered to exclude network prefixes that are known to be legitimate.'
This query can be filtered to exclude network prefixes that are known to be legitimate.'
Page 72 of 91