Threats Tagged 'xmrig'

An exposed directory at 188.245.99.156 revealed a comprehensive cryptomining toolkit containing 147 files including Python exploit source code, campaign logs, and Windows registry hives. Analysis confirms 3,562 Redis servers were compromised across two campaign runs targeting 12,966 hosts. The operation exploited unauthenticated Redis instances using rogue replication techniques to inject cron jobs that deployed XMRig miners. Victims spanned Redis versions 2.8.17 through 7.2.0 across outdated and current Linux distributions, indicating misconfiguration rather than version-specific vulnerabilities. The toolkit also targeted WordPress, MongoDB, and SSH but achieved zero confirmed compromises through those vectors. A separate February 2026 open directory linked by wallet reuse revealed Meterpreter deployment capabilities, extending the operator's known activity timeline by five months. The operation mined Monero through pool.moneroocean.stream with the same wallet used on the operator's own Windows-based work...

An emerging infostealer tracked as REVSTEALER under REF2859 has gained significant momentum with approximately 4,700 samples identified over the past year. The malware features comprehensive credential harvesting capabilities targeting browsers, VPN applications, password managers, cryptocurrency wallets, and gaming platforms. It incorporates sophisticated anti-analysis mechanisms including a weighted sandbox scoring system and uses Polygon blockchain-based dead drops for resilient infrastructure management. Distribution occurs primarily through social engineering campaigns targeting gamers via compromised YouTube channels advertising fake game cheats, though samples also impersonate legitimate software like Slack and qBittorrent. The malware delivers four additional modules extending capabilities to include wallet theft, clipboard manipulation, reverse proxy functionality, and cryptocurrency mining deployment. Most samples employ VMProtect packing and feature an App-Bound Encryption bypass using debugger-...

Multiple organizations experienced attacks beginning with social engineering that led to rogue ScreenConnect installations executing suspicious VBScript files. The attack chain involved four sequential VBScript payloads (1.vbs through 4.vbs) used to profile systems, check for security products, establish persistence, and deploy additional tools. Modified ScreenConnect clients enabled worm-like propagation by automatically transferring and executing these scripts on newly connected endpoints. The attacks included deployment of additional RMM tools like UltraViewer, cryptocurrency miners, and tunneling utilities. Systems were profiled based on installed security products, RAM capacity, and existing ScreenConnect installations. Persistence was achieved through Windows Registry Run Keys and concealed services. The campaign demonstrated sophisticated evasion techniques including AMSI bypass attempts, Windows Defender exclusions, and UAC bypass mechanisms.

In this article AI workloads are becoming high-value control points Case study 1: LiteLLM gateway compromise Case study 2: RAGFlow compromise Case study 3: Kestra compromise Mitigation and protection guidance MITRE ATT&CK techniques observed References Learn more AI is creating a new layer of enterprise infrastructure. Gateways, retrieval platforms, orchestration services, and containerized runtimes now sit between users, applications, data, and models. These systems concentrate credentials, data access, model connectivity, and execution privileges, making them some of the most powerful components in the AI stack. That concentration of trust is also creating new opportunities for attackers. In recent investigations, Microsoft observed activity targeting three distinct AI workloads: a LiteLLM gateway, a RAGFlow deployment, and a Kestra workflow environment. The intrusion paths varied, but the objectives were strikingly similar. Attackers sought to steal credentials, establish persistence, and monetize compromised compute resources. The individual techniques matter, but the broader pattern matters more. Across these cases, attackers treated AI infrastructure as a control plane where credential theft, host compromise, and downstream data access can converge. As organizations continue to deploy AI systems, these platforms are becoming high value targets that deserve the same security scrutiny as other critical enterprise infrastructure. AI workloads are becoming high-value control points The campaign-level signal extends beyond one product. The targeted workloads served different functions, but each exposed assets that could support follow-on abuse, including model-provider keys, proxy-issued virtual keys, database connection strings, tenant configuration, workflow execution, or host compute. Post-compromise behavior varied by workload role. Defenders should inventory exposed AI management surfaces, restrict administrative access, and monitor for gateway-originated execution and secret access. Three observed compromises across AI workloads AI workload Observed activity Attacker objective LiteLLM Observed attacker activity : Python droppers, runtime secret harvesting, PostgreSQL collection, miner deployment, and persistence activity from the LiteLLM gateway context. Microsoft assessment: Initial access likely occurred through exploitation of the exposed LiteLLM gateway surface, consistent with the vulnerability chain involving CVE-2026-42271 and CVE-2026-48710. Credential theft, backend database access, durable host access, and compute monetization. RAGFlow Observed attacker activity : Possible SSRF-style reconnaissance followed several days later by code execution, application-path modification, and placement of a Python hook in the TenantLLM credential-configuration flow. Public research: Describes multiple RAGFlow execution paths; Microsoft does not attribute this intrusion to a specific vulnerability. Intercept newly configured LLM provider credentials and model metadata. Kestra Observed attacker activity : Workflow-origin shell execution, Docker and container-environment discovery, XMRig deployment, and follow-on data collection. Microsoft assessment: Initial access likely involved exploitation of the exposed Kestra orchestration surface, with CVE-2026-49869 providing relevant public vulnerability context. Secret discovery, container-level access, data collection, and rapid compute monetization. Case study 1: LiteLLM gateway compromise Framework role and affected runtime context LiteLLM is commonly deployed as a proxy or gateway between applications and model providers. In that position, the service may hold or retrieve model-provider keys, LiteLLM master keys, virtual-key records, database connection strings, routing configuration, and tenant policy data. Command execution in the gateway runtime therefore exposed a process context close to AI routing and credential material. Figure 1. LiteLLM gateway compromise – attack chain…

This macOS malware campaign uses social engineering via fake CAPTCHA pages to trick users into running malicious AppleScript commands. It deploys a persistent backdoor that leverages EtherHiding by storing command-and-control (C2) addresses in Polygon blockchain smart contracts, complicating detection. The malware maintains persistence through LaunchAgents and delivers multiple payloads including the AMOS stealer, which targets cryptocurrency wallets, browser credentials, and macOS Keychain data, as well as the XMRig cryptominer for ongoing revenue. The campaign employs advanced evasion techniques such as character-ID obfuscation and abuse of legitimate macOS utilities. Blockchain transaction analysis reveals the full history of C2 infrastructure rotation and funding, aiding defenders despite the malware's memory-resident nature.

A sophisticated backdoor disguised as a legitimate Malwarebytes installer was distributed to over 100,000 machines through compromised automatic updates of one torrent client (Download Studio) and three adblockers (NetShield Kit, My AdBlock, and Net AdBlock). The backdoor creates a fake Malwarebytes installation directory containing legitimate signed files alongside malicious DLL files. Once executed, it establishes persistence through a Windows service and communicates with command-and-control servers to receive configuration updates and additional payloads. The primary observed payloads were cryptocurrency miners, though the infrastructure supports delivery of multiple persistent threats. The attack demonstrates abuse of software update mechanisms and affects primarily users in Russia, Ukraine, and Kazakhstan.

MediumMalwareKazakhstanRussiaUkraine#download studio#cryptocurrency mining#software update abuse

Aeternum is a C++ botnet loader utilizing the Polygon blockchain for command-and-control infrastructure instead of traditional centralized servers. Threat actors write encrypted and plaintext instructions directly to smart contracts, which infected devices query via public RPC endpoints. The malware implements weak PBKDF2HMAC/AES-GCM encryption with self-salting passwords, allowing payload decryption using only the smart contract address. Analysis reveals three related samples: the core Aeternum loader with Telegram-based exfiltration, a blended threat combining XWorm RAT with XMRig cryptocurrency miner, and Python source code revealing anti-analysis checks and cryptocurrency wallet targeting. The botnet demonstrates resilience through decentralized infrastructure, making traditional law enforcement takedowns significantly more challenging while maintaining low operational costs for attackers.

In May 2026, a sophisticated Monero cryptomining campaign was identified targeting Linux environments. Attackers gained initial access through trusted third-party relationships, then escalated to root privileges. Rather than operating openly as root, they weaponized Linux Pluggable Authentication Modules (PAM) to impersonate multiple low-privileged users, creating a forensic smokescreen and establishing redundant persistence through cronjobs. The operators suppressed system logging and deployed a customized XMRig 6.25.0 implant that self-unlinks after execution, running entirely in memory. The binary uses XOR encryption for configuration obfuscation and employs process masquerading to blend with legitimate processes. Campaign tracking revealed operations linked to the V25 Generation 26 family, connecting to the domain unable.download for mining pool communication.

Threat actors gaining initial access through SQL injection exploited a web application vulnerability in a technology sector organization. After compromising an MSSQL instance via inadequate input validation, the attackers deployed base64-encoded PowerShell scripts to conduct reconnaissance using tasklist commands and exfiltrated results to an external server. They established persistence by enabling Remote Desktop Services, creating an administratively privileged user account named adminweb2$, and disabling Windows Defender. The attackers installed BadIIS modules for SEO fraud, deployed XMRig cryptocurrency miner with hidden file attributes, and utilized service creation tools. Multiple PowerShell scripts and batch files were downloaded throughout the attack to facilitate various malicious operations and maintain access.

A malicious Go module posing as a DNS/subdomain scanner exposed a sophisticated Windows malware staging operation utilizing commit-farming workflows, public dead drops, and protected archives to deploy RAT and infostealer malware. The operation, tracked as 'Muck and Load', leverages a GitHub-based infrastructure comprising 222 confirmed repositories across 190 accounts designed to appear active and legitimate through automated GitHub Actions workflows. The attack chain begins with a deceptive Go module that downloads encoded PowerShell content, which then queries multiple public platforms including Pastebin, Telegram, YouTube, and Instagram as dead drops for encrypted payload locations. The loader retrieves password-protected archives containing AsyncRAT, Quasar, Remcos, and Vidar infostealer payloads, executing them from masqueraded Microsoft-themed directories. At least 14 malware files were confirmed across the repository network.