Remote WMI Process Execution
#
This Playbook is part of the Cortex Response And Remediation Pack.Supported versions
Supported Cortex XSOAR versions: 8.0.0 and later.
This playbook addresses the following alerts:
- Remote WMI process execution
- Suspicious remote WMI process execution
Playbook Stages:
Enrichment:
Enrich the attacker’s IP address to identify any known malicious activity.
Retrieve all incident-related alerts to consolidate context for further analysis.
Investigation:
Analyze command-line activity to assess risks based on suspicious patterns.
Check for high-confidence evidence, such as malicious IP addresses or suspicious command-line activity, to determine the next course of action.
Evaluate medium-confidence detections and request analyst approval for further containment if required.
Containment:
Attempt to terminate the malicious process tree using its causality ID.
Provide guidance for manual process termination if the automated action fails.
Propose endpoint isolation to prevent further compromise if malicious activity is confirmed.
#
DependenciesThis playbook uses the following sub-playbooks, integrations, and scripts.
#
Sub-playbooksThis playbook does not use any sub-playbooks.
#
IntegrationsCortexCoreIR
#
Scripts- CommandLineAnalysis
- SearchIncidentsV2
#
Commands- closeInvestigation
- core-get-cloud-original-alerts
- core-get-endpoints
- core-isolate-endpoint
- core-terminate-causality
- ip
#
Playbook InputsThere are no inputs for this playbook.
#
Playbook OutputsThere are no outputs for this playbook.