Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
40,407 detections
Filters
Last updated
All Time
Detection languages
15,618
13,637
2,824
1,803
1,727
Contributors
7,678
5,999
5,315
4,504
4,206
Categories
18,287
9,897
3,918
3,796
3,721
Platforms
40,407
7,473
6,580
3,935
3,713
Products / Services
10,635
9,722
6,762
1,902
1,895
MITRE Techniques
13,801
13,234
8,133
6,014
4,458
CVEs
49
45
30
30
29
IDS Classtypes
214
56
43
23
19
IDS Protocols
183
172
20
16
8
Detects a specific multi-stage brute-force execution chain involving a VBScript script executing batch files, searching result files via findstr.exe, and launching a fake dwm.exe process with specific command-line arguments typical of custom credential brute-forcing tools.
The following detects the execution of cscript.exe or wscript.exe processes spawned by cmd.exe, leveraging Endpoint Detection and Response (EDR) telemetry mapped to the Endpoint data model, with additional contextual filtering to improve fidelity and reduce false positives.
It focuses on executions originating from user-writable directories such as Users, AppData, Temp, and Downloads, which are commonly abused by attackers to stage and execute malicious scripts, while excluding trusted system paths like C:\Windows\System32\ and C:\Program Files\ that are typically associated with legitimate activity.
The detection also filters out service accounts (e.g., accounts ending with $ or known naming conventions) to minimize noise from automated processes and incorporates command-line context to better assess script execution patterns and identify potentially suspicious behavior.
It focuses on executions originating from user-writable directories such as Users, AppData, Temp, and Downloads, which are commonly abused by attackers to stage and execute malicious scripts, while excluding trusted system paths like C:\Windows\System32\ and C:\Program Files\ that are typically associated with legitimate activity.
The detection also filters out service accounts (e.g., accounts ending with $ or known naming conventions) to minimize noise from automated processes and incorporates command-line context to better assess script execution patterns and identify potentially suspicious behavior.
Detects indicators of compromise (IOCs) and suspicious behaviors associated with the BITTER threat group's PapRAT and WmRAT campaigns on Windows systems, including specific file hashes, malicious domains, C2 IP addresses, dropped files, scheduled tasks, and anomalous process execution chains.
Detects potential DLL sideloading of CertEnroll.dll via CertEnrollCtrl.exe spawned from non-standard paths (such as ProgramData, AppData, or Temp) and associated persistence mechanisms using batch files to create scheduled tasks or registry run keys.
Detects Microsoft Edge used as a hidden HTTP client by the HOOKEDGE backdoor (APT28 / Swallowtail) an inline data:text/html URL in a 1x1 or off-screen window, or a script-launched Edge with an inline data:text/html URL and --ignore-certificate-errors. In the observed intrusion HOOKEDGE was followed by hands-on-keyboard credential theft, an SSH reverse tunnel and a reflective loader (ld.exe), so a hit should be treated as a potential active compromise rather than an isolated alert.
Detects the distinctive operational loop of autonomous AI attack agents:
repeated cycles of enumeration activity followed by exploitation attempts
followed by further enumeration — all targeting the same host or subnet.
This cyclical pattern reflects the planner-worker architecture where
exploitation results feed back into the exploration graph, triggering new
rounds of discovery.
Human operators may enumerate, exploit, and move on. Autonomous frameworks
like ARTEX loop: scan, attempt exploit, re-scan to discover what changed
(new access, new services exposed, new credentials available), attempt
more exploits with updated context, re-scan again. Each cycle produces
a burst of network discovery events (port scans, service probes) followed
by authenticated access attempts, then another burst of discovery from the
newly accessed host — repeating 3+ times within a session.
This detection requires correlation of network connection events with
authentication events on the same host or subnet, tracking the oscillation
between discovery and access phases.
BQL (Bloo Query Language):
stream=ep-network where action='NETWORK_CONNECTION'
and DstPort in (445, 135, 139, 22, 3389, 5985, 5986, 389, 636, 88)
and DstIP rlike '^(10\.|192\.168\.|172\.(1[6-9]|2[0-9]|3[01])\.)'
| groupby system, DstIP
| having count > 15
repeated cycles of enumeration activity followed by exploitation attempts
followed by further enumeration — all targeting the same host or subnet.
This cyclical pattern reflects the planner-worker architecture where
exploitation results feed back into the exploration graph, triggering new
rounds of discovery.
Human operators may enumerate, exploit, and move on. Autonomous frameworks
like ARTEX loop: scan, attempt exploit, re-scan to discover what changed
(new access, new services exposed, new credentials available), attempt
more exploits with updated context, re-scan again. Each cycle produces
a burst of network discovery events (port scans, service probes) followed
by authenticated access attempts, then another burst of discovery from the
newly accessed host — repeating 3+ times within a session.
This detection requires correlation of network connection events with
authentication events on the same host or subnet, tracking the oscillation
between discovery and access phases.
BQL (Bloo Query Language):
stream=ep-network where action='NETWORK_CONNECTION'
and DstPort in (445, 135, 139, 22, 3389, 5985, 5986, 389, 636, 88)
and DstIP rlike '^(10\.|192\.168\.|172\.(1[6-9]|2[0-9]|3[01])\.)'
| groupby system, DstIP
| having count > 15
Detects a process reading SSH private key files (.ssh/id_rsa, id_ed25519,
authorized_keys) followed by outbound SSH connections to previously
unseen internal hosts. Autonomous AI agents harvest SSH credentials from
compromised hosts and immediately reuse them for lateral movement — the
time between key discovery and first use is measured in seconds.
ARTEX worker agents performing credential harvesting scan predictable
paths (/home/*/.ssh/, /root/.ssh/, C:\Users\*\.ssh\) and feed discovered
keys directly into the exploration graph. The planner then dispatches
workers to authenticate to every host in the asset graph using the
harvested keys. The compressed window between file read and SSH connection
is the differentiator.
Requires Sysmon Event ID 1 (process creation) and Event ID 3 (network
connection) with process-level correlation.
BQL (Bloo Query Language):
stream=ep-process where action='PROCESS_ADDED'
and (lower(commandline) like '%.ssh/id_rsa%'
or lower(commandline) like '%.ssh/id_ed25519%'
or lower(commandline) like '%authorized_keys%'
or lower(commandline) like '%.ssh\\id_rsa%')
| groupby system, user
authorized_keys) followed by outbound SSH connections to previously
unseen internal hosts. Autonomous AI agents harvest SSH credentials from
compromised hosts and immediately reuse them for lateral movement — the
time between key discovery and first use is measured in seconds.
ARTEX worker agents performing credential harvesting scan predictable
paths (/home/*/.ssh/, /root/.ssh/, C:\Users\*\.ssh\) and feed discovered
keys directly into the exploration graph. The planner then dispatches
workers to authenticate to every host in the asset graph using the
harvested keys. The compressed window between file read and SSH connection
is the differentiator.
Requires Sysmon Event ID 1 (process creation) and Event ID 3 (network
connection) with process-level correlation.
BQL (Bloo Query Language):
stream=ep-process where action='PROCESS_ADDED'
and (lower(commandline) like '%.ssh/id_rsa%'
or lower(commandline) like '%.ssh/id_ed25519%'
or lower(commandline) like '%authorized_keys%'
or lower(commandline) like '%.ssh\\id_rsa%')
| groupby system, user
This rule performs an IOC sweep across multiple telemetry sources (DeviceNetworkEvents, DeviceEvents, DeviceFileEvents, CommonSecurityLog, DnsEvents) to detect network connections, DNS queries, and file activity associated with the Optimizely supply chain attack distributing Coruna/DarkSword malware.
This rule performs an IOC sweep across multiple telemetry sources (DeviceNetworkEvents, DeviceEvents, DeviceFileEvents, CommonSecurityLog, DnsEvents) to detect network connections, DNS queries, and file activity associated with the Optimizely supply chain attack distributing Coruna/DarkSword malware.
This rule performs an IOC sweep across multiple telemetry sources (DeviceNetworkEvents, DeviceEvents, DeviceFileEvents, CommonSecurityLog, DnsEvents) to detect network connections, DNS queries, and file activity associated with the Optimizely supply chain attack distributing Coruna/DarkSword malware.
Detects the execution or creation of the remote control utility rs.win32.exe from non-standard directories like AppData\Roaming\CMS\RemoteSetup with specific command-line parameters or when launched by CMS.exe.
CVE-2026-83844 is an out-of-bounds read remote code execution vulnerability in Windows Paint (mspaint.exe). This CrowdStrike Falcon CQL query works as both a detection and a threat hunting query. It surfaces suspicious mspaint.exe behavior that may indicate exploitation. Paint is an image editor and has little legitimate reason to launch programs, write executables or scripts, or connect to remote hosts.
How it detects
The query reads Windows endpoint telemetry from ProcessRollup2, NewExecutableWritten, NewScriptWritten, DnsRequest, NetworkConnectIP4 and NetworkConnectIP6 events. It matches four behaviors:
1. Process lineage (hunt): Any process where mspaint.exe is the parent (ParentBaseFileName). This includes scripting hosts such as cmd.exe, powershell.exe and wscript.exe, LOLBins, and network-capable binaries. Paint should not spawn child processes.
2. Executable or script written to disk (alert): mspaint.exe, as the context process, creating executable or script content (NewExecutableWritten, NewScriptWritten).
3. DNS lookups (alert): Domain resolution requests made by mspaint.exe (DnsRequest).
4. Outbound network connections (alert): IPv4 and IPv6 connections made by mspaint.exe (NetworkConnectIP4, NetworkConnectIP6). The two address fields are combined into one RemoteAddress field.
Known benign activity
mspaint.exe may contact "apsaiservices-a0fqcjc6bzbhgdcd.b02.azurefd.net" and "150.171.110.85" for telemetry. The query does not exclude these destinations, so they will appear as DNS and network events. Treat them as likely benign, but still check each hit in context. Other false positives may occur, such as legitimate Paint integrations and updates.
Triage guidance
Identify the file that opened Paint, such as an email attachment or a download. Review any child process command line and the hash of any dropped file. Check unfamiliar destinations and isolate the host if the activity is confirmed malicious.
How it detects
The query reads Windows endpoint telemetry from ProcessRollup2, NewExecutableWritten, NewScriptWritten, DnsRequest, NetworkConnectIP4 and NetworkConnectIP6 events. It matches four behaviors:
1. Process lineage (hunt): Any process where mspaint.exe is the parent (ParentBaseFileName). This includes scripting hosts such as cmd.exe, powershell.exe and wscript.exe, LOLBins, and network-capable binaries. Paint should not spawn child processes.
2. Executable or script written to disk (alert): mspaint.exe, as the context process, creating executable or script content (NewExecutableWritten, NewScriptWritten).
3. DNS lookups (alert): Domain resolution requests made by mspaint.exe (DnsRequest).
4. Outbound network connections (alert): IPv4 and IPv6 connections made by mspaint.exe (NetworkConnectIP4, NetworkConnectIP6). The two address fields are combined into one RemoteAddress field.
Known benign activity
mspaint.exe may contact "apsaiservices-a0fqcjc6bzbhgdcd.b02.azurefd.net" and "150.171.110.85" for telemetry. The query does not exclude these destinations, so they will appear as DNS and network events. Treat them as likely benign, but still check each hit in context. Other false positives may occur, such as legitimate Paint integrations and updates.
Triage guidance
Identify the file that opened Paint, such as an email attachment or a download. Review any child process command line and the hash of any dropped file. Check unfamiliar destinations and isolate the host if the activity is confirmed malicious.
Detects registry writes to browser extension policy keys (ExtensionInstallForcelist, ExtensionInstallAllowlist, ExtensionSettings) for Google Chrome or Microsoft Edge by processes not explicitly approved as standard management tools.
Detects indicators of compromise (IOCs) and suspicious behaviors associated with the BITTER threat group's PapRAT and WmRAT campaigns on Windows systems, including specific file hashes, malicious domains, C2 IP addresses, dropped files, scheduled tasks, and anomalous process execution chains.
Detects wscript.exe or cscript.exe deleting the Zone.Identifier alternate data stream (Mark-of-the-Web) of a .js file in the Downloads folder, commonly observed in obfuscated script-based loaders attempting to bypass security controls.
Detects rare, non-system, non-browser processes opening a TCP listening connection on port 56793, which is associated with the rat_http3_cc embedded local HTTP server. Excludes processes that have a baseline of being seen listening on this port across multiple devices.
This detection identifies potential FakeCAPTCHA/ClickFix clipboard hijacking campaigns by looking for PowerShell execution with hidden window parameters and distinctive strings related to fake CAPTCHA verification. These campaigns use social engineering to trick users into pasting malicious PowerShell commands from their clipboard, typically delivering information stealers or remote access trojans.
Outbound connections from finger.exe
Cortex XDR
Detects network activity associated with the execution of the finger.exe utility, which can be used to query user information and perform network reconnaissance.
Detects curl command lines performing file uploads or pushing data over FTP/SFTP, especially when combined with proxy or encrypted-channel options, which may indicate data exfiltration.
Detects a sequence where WpcMon.exe is created or modified within a user AppData Local Microsoft WpcMon path, followed by a Registry Run key value creation pointing to it within 1 hour on the same device, associated with the CMS.exe compromise campaign.
Detects a sequence of events related to malicious scheduled task activity involving a kvm-agent artifact: scheduled task creation using schtasks.exe, command-line process discovery using tasklist to check for kvm-agent.exe, and execution of kvm-agent.exe from the SysWOW64 directory.
Page 9 of 1925