SMOKE#SCREEN Abuses ScreenConnect and Cloudflare Tunnels
Detection stack
- AIDR
- Alert
- ETL
- Query
Summary
The SMOKE#SCREEN campaign involves threat actors deploying legitimate ScreenConnect RMM agents through convincing social engineering lures disguised as Zoom and Adobe updates. The attackers rely on a rotating toolkit of VBScript droppers, .NET loaders, and batch files to bypass security controls and maintain persistent remote access. The campaign reflects evolving tradecraft, shifting from aggressive security software disruption toward stealthier techniques timed to evade EDR detection.
Investigation
Securonix Threat Research uncovered a multi-wave campaign that used WsgiDAV staging servers and Cloudflare tunnels to deliver payloads. Investigators identified three distinct ScreenConnect relay clusters and reconstructed five separate kill chains by analyzing multiple scripts and executable samples. The research also showed a progression from heavily obfuscated VBScript to advanced compiled loaders designed to operate outside specific EDR detection windows.
Mitigation
Defenders should restrict untrusted MSI package execution through AppLocker or WDAC and monitor for unauthorized tampering with security products. Auditing legitimate RMM tool usage and enforcing strict UAC policies are essential for preventing unauthorized remote access. Behavioral EDR rules should also be deployed to identify suspicious parent-child process relationships and unusual installer activity.
Response
If SMOKE#SCREEN activity is detected, affected systems should be isolated immediately to terminate active ScreenConnect sessions. Responders should investigate the original access vector, including phishing emails and malicious software downloads. Any exposed credentials should be revoked, while system configuration changes, particularly Microsoft Defender exclusions and registry modifications, should be reviewed and reversed.
Attack Flow
Detections
Disabling Windows Defender Protections (via registry_event)
Proof of Value
Suspicious Ransomware Interfering Service Stoppage (via cmdline)
Proof of Value
Download or Upload via Powershell (via cmdline)
Proof of Value
Call Suspicious .NET Methods from Powershell (via powershell)
Proof of Value
Windows Defender Preferences Suspicious Changes (via powershell)
Proof of Value
Possible Defense Evasion by AMSI Bypass Attempt (via powershell)
Proof of Value
Possible RMM Software Installation Attempt Using MsiInstaller (via application logs)
Proof of Value
Suspicious File Download Direct IP (via proxy)
Proof of Value
Suspicious Trycloudflare Domain Communication (via dns)
Proof of Value
IOCs (HashSha256) to detect: Analyzing SMOKE#SCREEN: ScreenConnect RMM Abuse, Cloudflare Tunnels, and Trusted Software Lures
Proof of Value
IOCs (SourceIP) to detect: Analyzing SMOKE#SCREEN: ScreenConnect RMM Abuse, Cloudflare Tunnels, and Trusted Software Lures
Proof of Value
IOCs (DestinationIP) to detect: Analyzing SMOKE#SCREEN: ScreenConnect RMM Abuse, Cloudflare Tunnels, and Trusted Software Lures
Proof of Value
Detection of Malicious Process Creation Techniques [Windows Process Creation]
Proof of Value
Windows Defender and SmartScreen Tampering via Registry Modification [Windows Registry Event]
Proof of Value
PowerShell In-Memory C# Execution and Security Service Disabling [Windows Powershell]
Proof of Value
Silent Install of ScreenConnect MSI Indicating Potential MSI-based Attack [Windows Sysmon]
Proof of Value
Simulation Execution
Prerequisite: The Telemetry & Baseline Pre-flight Check must have passed.
Rationale: This section details the precise execution of the adversary technique (TTP) designed to trigger the detection rule. The commands and narrative MUST directly reflect the TTPs identified and aim to generate the exact telemetry expected by the detection logic. Abstract or unrelated examples will lead to misdiagnosis.
-
Attack Narrative & Commands: The adversary has gained initial access and seeks to deploy ScreenConnect for persistent remote access. To avoid user interruption and detection, the attacker uses
msiexec.exewith the/qn(quiet, no UI) flag to install a malicious MSI package. Once the installation is initiated, the attacker attempts to neutralize the host’s primary defense by using PowerShell to add a Windows Defender exclusion for their tool directory and then attempts to stop the Windows Defender service entirely to prevent real-time scanning of their subsequent toolkit. -
Regression Test Script:
# Simulation Script: ScreenConnect Silent Install & Defense Evasion # 1. Create a dummy MSI file for simulation purposes (simulating the ScreenConnect installer) # In a real test, a valid MSI would be used. $msiPath = "$env:TEMPScreenConnect_Sim.msi" Write-Host "[+] Creating dummy MSI at $msiPath" # Note: This is a placeholder logic. For true telemetry, a real MSI must be present. # 2. Execute Silent MSI Installation (Trigger: selection_processname_msi) Write-Host "[+] Executing silent MSI installation..." Start-Process msiexec.exe -ArgumentList "/i $msiPath /qn" -Wait # 3. PowerShell Defense Evasion - Add Exclusion (Trigger: selection_powershell_exclusion) Write-Host "[+] Attempting to add Defender exclusion via PowerShell..." powershell.exe -Command "Add-MpPreference -ExclusionPath 'C:Program FilesScreenConnect'" # 4. PowerShell Defense Evasion - Stop Defender Service (Trigger: selection_stop_defender) Write-Host "[+] Attempting to stop Windows Defender service..." powershell.exe -Command "Stop-Service -Name WinDefend" -
Cleanup Commands:
# Cleanup Simulation Artifacts Remove-Item -Path "$env:TEMPScreenConnect_Sim.msi" -ErrorAction SilentlyContinue # Note: Reversing Defender changes requires administrative privileges and may impact security posture. # Only perform if in a strictly isolated lab environment. Write-Host "[!] Cleanup complete. Manual restoration of Defender settings recommended."