Detection rules › Splunk

Windows SSH Proxy Command

Status
production
Severity
low
Group by
IntegrityLevel, command_line, computer_name, event_action, original_file_name, parent_command_line, parent_process_guid, parent_process_id, parent_process_name, process_guid, process_hash, process_id, process_name, user, user_id, vendor_product
Author
Michael Haag, AJ King, Nasreddine Bencherchali, Splunk, Jesse Hunter, Splunk Community Contributor
Source
github.com/splunk/security_content

This detection identifies potential abuse of SSH "ProxyCommand" or "LocalCommand" by monitoring for suspicious process execution patterns. Specifically, it looks for instances where ssh.exe (as a parent process) containing "ProxyCommand" or "LocalCommand" in its arguments spawns potentially malicious child processes like mshta, powershell, wscript, or cscript, or processes containing "http" in their command line. This technique can be used by attackers to execute arbitrary commands through SSH proxy configurations, potentially enabling command & control activities or remote code execution. The detection focuses on commonly abused Windows scripting engines and web requests that may indicate malicious activity when spawned through SSH proxy commands.

MITRE ATT&CK coverage

Event coverage

ProviderEventTitle
SysmonEvent ID 1Process creation

Rule body splunk

name: Windows SSH Proxy Command
id: ac520039-21f1-4567-b528-5b7133dba76f
version: 6
creation_date: '2025-03-24'
modification_date: '2026-05-13'
author: Michael Haag, AJ King, Nasreddine Bencherchali, Splunk, Jesse Hunter, Splunk Community Contributor
status: production
type: Anomaly
description: |
    This detection identifies potential abuse of SSH "ProxyCommand" or "LocalCommand" by monitoring for suspicious process execution patterns.
    Specifically, it looks for instances where ssh.exe (as a parent process) containing "ProxyCommand" or "LocalCommand" in its arguments spawns potentially malicious child processes like mshta, powershell, wscript, or cscript, or processes containing "http" in their command line.
    This technique can be used by attackers to execute arbitrary commands through SSH proxy configurations, potentially enabling command & control activities or remote code execution. The detection focuses on commonly abused Windows scripting engines and web requests that may indicate malicious activity when spawned through SSH proxy commands.
data_source:
    - Sysmon EventID 1
    - CrowdStrike ProcessRollup2
search: |-
    | tstats `security_content_summariesonly`
      count min(_time) as firstTime
            max(_time) as lastTime
    
    from datamodel=Endpoint.Processes where
    
    Processes.parent_process_name="ssh.exe"
    (
      Processes.parent_process = "*ProxyCommand=*"
      OR
      (
        Processes.parent_process = "* PermitLocalCommand=yes*"
        Processes.parent_process = "* LocalCommand=*"
      )
    )
    
    Processes.process IN (
      "*cscript*",
      "*http*",
      "*mshta*",
      "*powershell*",
      "*pwsh*",
      "*wmic*",
      "*wscript*"
    )
    
    by Processes.action Processes.dest Processes.original_file_name Processes.parent_process Processes.parent_process_exec
       Processes.parent_process_guid Processes.parent_process_id Processes.parent_process_name Processes.parent_process_path
       Processes.process Processes.process_exec Processes.process_guid Processes.process_hash Processes.process_id
       Processes.process_integrity_level Processes.process_name Processes.process_path Processes.user Processes.user_id
       Processes.vendor_product
    
    | `drop_dm_object_name(Processes)`
    | `security_content_ctime(firstTime)`
    | `security_content_ctime(lastTime)`
    | `windows_ssh_proxy_command_filter`
how_to_implement: |
    The detection is based on data that originates from Endpoint Detection and Response (EDR) agents. These agents are designed to provide security-related telemetry from the endpoints where the agent is installed. To implement this search, you must ingest logs that contain the process GUID, process name, and parent process. Additionally, you must ingest complete command-line executions. These logs must be processed using the appropriate Splunk Technology Add-ons that are specific to the EDR product. The logs must also be mapped to the `Processes` node of the `Endpoint` data model. Use the Splunk Common Information Model (CIM) to normalize the field names and speed up the data modeling process.
known_false_positives: |
    Legitimate use of SSH ProxyCommand or LocalCommand with scripting engines may trigger this detection.
    Filter as needed based on your environment's normal SSH usage patterns and authorized scripting activities.
references:
    - https://www.virustotal.com/gui/file/c33f82868dbbfc3ab03918f430b1a348499f5baf047b136ff0a4fc3e8addaa9b/detection
    - https://attack.mitre.org/techniques/T1572/
    - https://lolbas-project.github.io/lolbas/Binaries/Ssh/
    - https://man.openbsd.org/ssh_config#ProxyCommand
    - https://man.openbsd.org/ssh_config#LocalCommand
drilldown_searches:
    - name: View the detection results for - "$user$" and "$dest$"
      search: '%original_detection_search% | search user = "$user$" dest = "$dest$"'
      earliest_offset: $info_min_time$
      latest_offset: $info_max_time$
    - name: View risk events for the last 7 days for - "$user$" and "$dest$"
      search: '| from datamodel Risk.All_Risk | search normalized_risk_object IN ("$user$", "$dest$") | stats count min(_time) as firstTime max(_time) as lastTime values(search_name) as "Search Name" values(risk_message) as "Risk Message" values(analyticstories) as "Analytic Stories" values(annotations._all) as "Annotations" values(annotations.mitre_attack.mitre_tactic) as "ATT&CK Tactics" by normalized_risk_object | `security_content_ctime(firstTime)` | `security_content_ctime(lastTime)`'
      earliest_offset: 7d
      latest_offset: "0"
intermediate_findings:
    entities:
        - field: dest
          type: system
          score: 20
          message: Suspicious process execution $process$ detected through SSH $parent_process$ on $dest$ by user $user$
        - field: user
          type: user
          score: 20
          message: Suspicious process execution $process$ detected through SSH $parent_process$ on $dest$ by user $user$
threat_objects:
    - field: parent_process_name
      type: parent_process_name
    - field: process_name
      type: process_name
analytic_story:
    - ZDI-CAN-25373 Windows Shortcut Exploit Abused as Zero-Day
    - Living Off The Land
    - Hellcat Ransomware
asset_type: Endpoint
mitre_attack_id:
    - T1572
    - T1059.001
    - T1105
product:
    - Splunk Enterprise
    - Splunk Enterprise Security
    - Splunk Cloud
category: endpoint
security_domain: endpoint
tests:
    - name: True Positive Test
      attack_data:
        - data: https://media.githubusercontent.com/media/splunk/attack_data/master/datasets/attack_techniques/T1572/ssh_proxy_command/sshproxycommand_windows-sysmon.log
          sourcetype: XmlWinEventLog
          source: XmlWinEventLog:Microsoft-Windows-Sysmon/Operational
      test_type: unit

Stages and Predicates

Stage 1: tstats

| tstats `security_content_summariesonly`
  count min(_time) as firstTime
        max(_time) as lastTime

from datamodel=Endpoint.Processes where

Processes.parent_process_name="ssh.exe"
(
  Processes.parent_process = "*ProxyCommand=*"
  OR
  (
    Processes.parent_process = "* PermitLocalCommand=yes*"
    Processes.parent_process = "* LocalCommand=*"
  )
)

Processes.process IN (
  "*cscript*",
  "*http*",
  "*mshta*",
  "*powershell*",
  "*pwsh*",
  "*wmic*",
  "*wscript*"
)

by Processes.action Processes.dest Processes.original_file_name Processes.parent_process Processes.parent_process_exec
   Processes.parent_process_guid Processes.parent_process_id Processes.parent_process_name Processes.parent_process_path
   Processes.process Processes.process_exec Processes.process_guid Processes.process_hash Processes.process_id
   Processes.process_integrity_level Processes.process_name Processes.process_path Processes.user Processes.user_id
   Processes.vendor_product

Stage 2: search

| `drop_dm_object_name(Processes)`

Stage 3: search

| `security_content_ctime(firstTime)`

Stage 4: search

| `security_content_ctime(lastTime)`

Stage 5: search

| `windows_ssh_proxy_command_filter`

Indicators

Each row is a field, operator, and value that the rule matches. The corpus column counts how many other rules in the catalog look for the same combination: high numbers point to widely-used, community-vetted indicators. Blank or 1 shows that the indicator is specific to this rule.

FieldKindValues
Processes.parent_processeq
  • "* LocalCommand=*"
  • "* PermitLocalCommand=yes*"
  • "*ProxyCommand=*"
Processes.parent_process_nameeq
  • "ssh.exe"
Processes.processin
  • "*cscript*" corpus 15 (sigma 15)
  • "*http*" corpus 39 (sigma 34, elastic 2, chronicle 2, splunk 1)
  • "*mshta*" corpus 14 (sigma 14)
  • "*powershell*" corpus 25 (sigma 24, chronicle 1)
  • "*pwsh*" corpus 7 (sigma 7)
  • "*wmic*" corpus 6 (sigma 6)
  • "*wscript*" corpus 16 (sigma 16)