#threatreport #HighCompleteness
Velvet Ant’s Operation Highland: How a China-Nexus Actor Infiltrated an Internal Network Undetected | 11-06-2026
Source:
sygnia.co/blog/operation-hig…
Key details below ↓
🧑💻Actors/Campaigns:
Velvet_ant
Operation_highland
💀Threats:
Velvetshell, Gs-netcat, Credential_harvesting_technique, Credential_dumping_technique, Timestomp_technique, Logcleaner_tool, Pnscan_tool, Pscan_tool,
🎯Victims: Major organization, Critical infrastructure
🏭Industry: Critical_infrastructure, Healthcare
🌐Geo: China
📚TTPs:
⚔️Tactics: 4
🛠️Technics: 0
🧨IOCs:
- Domain: 9
- IP: 5
- Hash: 276
- Coin: 1
💽Software: OpenSSH, systemd, Chrome, Nginx, FastCGI, SELinux, ELinux di, Linux
🔢Algorithms: sha1, md5
🔠Functions: Internet-Facing
📜Programming Languages: perl
#threatreport:
The investigation by Sygnia into Velvet Ant’s Operation Highland revealed a sophisticated cyber attack attributed to a China-nexus threat actor, characterized by a nearly decade-long undetected presence within a critical infrastructure network. The initial compromise dates back to 2016, with Velvet Ant demonstrating advanced strategies for maintaining persistence and control over the network by targeting the authentication stack crucial for access management.
Velvet Ant executed a multi-stage attack strategy, starting with gaining access to internet-facing systems before pivoting through the IT network to infiltrate the isolated critical infrastructure segment. One of the paramount vulnerabilities exploited was the compromise of the Pluggable Authentication Module (PAM) and OpenSSH binaries, facilitating complete oversight over authentication processes. This manipulation allowed the threat actor to observe all credential usage, effectively sidestepping standard authentication checks, leading to persistent access that remained intact despite any password changes.
Among the specific technical details of Velvet Ant’s methods, the use of backdoored versions of PAM modules (such as pam_unix.so) and maliciously modified OpenSSH binaries revealed the extent of their operational capabilities. Nine variants of the pam_unix.so module were identified, each individually compiled and designed to bypass authentication by embedding backdoor passwords and harvesting legitimate credentials through tampered authentication flows. Additionally, altered OpenSSH binaries captured and stored user credentials and logged shell commands while including functionalities designed to disable logging and access control measures on the systems.
Velvet Ant’s arsenal also incorporated tools such as a modified version of GS-Netcat as a covert reverse shell, encrypted for stealth while the threat actor blended its presence within system processes by manipulating process names. Furthermore, a custom execution bridge using Nginx and FastCGI allowed for remote command execution through HTTP requests, enabling access to the critical infrastructure without necessitating a direct connection.
Remediation of the attack posed significant challenges due to the manipulative techniques employed by Velvet Ant. Merely replacing the compromised services would not suffice, as any incorrect removal of altered PAM modules or OpenSSH binaries could result in access denial for administrators, jeopardizing operational continuity. Therefore, a meticulous remediation strategy was developed to carefully replace malicious components, ensuring minimal disruption whilst validating SSH and authentication health through rigorous testing.
Ultimately, Operation Highland exemplifies the intricacies of maintaining undetected access in high-security environments, emphasizing the intricacies faced by defenders in detecting insidious attacks that exploit trusted system components. The ability of threat actors to integrate deeply within system processes necessitates proactive threat hunting and an analytical approach directed at identifying unexpected or inconsistent behaviors, rather than relying solely on signature-based detection.