Skip to main content
Live Webinar with SANS (June 25)— Agentic CTI Automation for Fun & ProfitRegister Free
Mallory
Back to malware
MalwareRansomwareExploits 3 CVEs

VSOCKpuppet

VSOCKpuppet is a Linux/ESXi 64-bit ELF backdoor deployed on compromised VMware ESXi hosts as part of a VM-escape exploit toolkit analyzed by Huntress. It is described as an ESXi-resident payload used after a guest-to-host escape chain likely involving CVE-2025-22226, CVE-2025-22224, and CVE-2025-22225. The broader toolkit includes the Windows orchestrator MAESTRO (exploit.exe), the unsigned kernel driver MyDriver.sys, and a Windows VSOCK client component referred to as client.exe or the GetShell Plugin.

According to the content, VSOCKpuppet provides persistent remote access to the ESXi host and communicates over VSOCK (Virtual Sockets), specifically listening on VSOCK port 10000 with VMADDR_CID_ANY. Its functionality includes arbitrary shell command execution via /bin/sh and file transfer operations, including reading and writing files (described as GET and POST support). The malware is characterized as stealthy because its guest-host communications over VSOCK can bypass or evade traditional network monitoring and IDS visibility.

The reported deployment flow writes VSOCKpuppet into VMX memory along with stage-1 and stage-2 shellcode. Stage-2 shellcode then writes the backdoor to /var/run/a, temporarily modifies /var/run/inetd.conf so inetd will execute /var/run/a as root via port 21, sends SIGHUP to inetd to reload the configuration, triggers the backdoor locally, and restores inetd.conf afterward to reduce detection. The content also states the attackers restored drivers and cleaned up configuration changes as part of a stealth-focused approach.

The malware was observed in intrusions investigated in December 2025 in which attackers likely gained initial access through a compromised SonicWall VPN appliance, used compromised Domain Admin credentials for lateral movement, and then deployed the ESXi exploit toolkit from Windows systems. Huntress assessed the operators as Chinese-speaking based on simplified Chinese development artifacts in related tooling, though the content does not attribute VSOCKpuppet itself to a specific named threat actor. The activity targeted VMware ESXi hypervisors and was associated with data staging for exfiltration and possible ransomware preparation.

High-confidence indicators directly mentioned in the content include the filename/name VSOCKpuppet, its classification as a Linux-based/ELF payload on ESXi, VSOCK port 10000, the dropped path /var/run/a, and the SHA-256 hash c3f8da7599468c11782c2332497b9e5013d98a1030034243dfed0cf072469c89.

Share:
For your environment

Hunt this family in your stack

Mallory pivots from this family to the IOCs, detections, and named campaigns that touch your stack, and pages you when something new lands.

EXPLOITED CVES

Vulnerabilities exploited

3 CVEs Mallory has correlated with this family across public research and vendor advisories. Each row links to the full Mallory page for that vulnerability.

3 CVES
CVE-2025-22225VMware ESXi VMX sandbox escape via arbitrary kernel writeExploited in the wild

CVE-2025-22225 (8.2 severity score): An arbitrary write vulnerability in ESXi that allows escaping the VMX sandbox to the kernel

via bleeping computerbleepingcomputer.com
CVE-2025-22224TOCTOU out-of-bounds write in VMware ESXi and Workstation VMCIExploited in the wild

CVE-2025-22224 (9.3 severity score): A TOCTOU vulnerability in Virtual Machine Communication Interface (VMCI) leading to an out-of-bounds write, allowing code execution as the VMX process

via bleeping computerbleepingcomputer.com
CVE-2025-22226Out-of-bounds Read in VMware HGFSExploited in the wild

Of the three bugs, only one received a critical severity score: CVE-2025-22226 (7.1 severity score): An out-of-bounds read in HGFS that allows leaking memory from the VMX process

via bleeping computerbleepingcomputer.com
MITRE ATT&CK

Techniques & procedures

2 distinct techniques documented for this family, organized by ATT&CK tactic.

Initial Access

1 technique
T1190Exploit Public-Facing ApplicationEvidence1

"Huntress assesses that initial access likely came through a compromised SonicWall VPN."

Exfiltration

1 technique
T1020Automated ExfiltrationEvidence1

"...stage data for exfiltration..."

What this page doesn’t show

The version that knows your environment.

This page is what’s public. Mallory adds the parts that aren’t: which of your assets match these IOCs, which detections are missing, which campaigns to expect next, and what to do in the next 30 minutes.
IOC matching

Match every observed IP, domain, and hash against your live telemetry.

Threat actor attribution

Named campaigns wielding this family, with evidence pinned to each claim.

Exploited vulnerabilities3

CVEs this family uses for access and lateral movement.

Detection signatures

YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.

MITRE ATT&CK mapping2

Every documented technique, ranked by evidence weight.

Researcher chatter

Reddit, Mastodon, and CTI community discussion around this family.