In June 2026, a compromised Klue competitive-intelligence platform integration was exploited to exfiltrate customer relationship management data from enterprise Salesforce environments. Attackers authenticated through compromised Klue service accounts, generated OAuth tokens, and executed automated Python scripts to conduct bulk data extraction via Salesforce REST API queries over approximately 24 hours. The activity included concentrated bursts of nearly a thousand queries within 15 minutes and sustained extraction windows exceeding 6 hours. This incident follows similar third-party OAuth-abuse campaigns targeting Salesforce through Salesloft Drift and Gainsight integrations throughout 2025 and 2026. While the tactics resemble operations attributed to ShinyHunters and UNC6395 threat groups, attribution remains uncertain. The initial access vector, full scope of exfiltration, and attacker intent are still under investigation, with no extortion demands observed to date.
NadMesh is an industrial-grade Go-based botnet observed in July 2026 that autonomously scans and exploits AI infrastructure and cloud services. The botnet integrates scanning, exploitation, and intelligence harvesting into a single platform targeting over 90 cloud provider address ranges. It employs 20+ exploitation vectors against Redis, Docker, MCP, Kubernetes, and other services, with particular focus on AI platforms like ComfyUI, Ollama, and Gradio discovered via Shodan API. NadMesh features a web-based management panel, polymorphic builds using Garble obfuscation and UPX packing, and redundant persistence mechanisms including SSH backdoors, agent processes, and cron watchdogs. The operation demonstrates clear commercial intent with conversion funnel statistics, canary updates, and automated task supply loops that amplify high-yield subnets. It harvests cloud credentials, Kubernetes tokens, AI model access, and MCP service intelligence.
A DPRK-aligned threat group is targeting developers through fake job postings and coding challenges in a campaign tracked as REF9403. Attackers post fake job offers in developer forums, then send trojanized repositories containing fully functional e-commerce projects with malicious code hidden using steganography inside SVG flag images. When developers run these projects, the malware deploys four-stage payloads aligned with OTTERCOOKIE: a browser credential and cryptocurrency wallet stealer, a file exfiltration module, a Socket.IO-based remote access trojan, and a clipboard stealer. The campaign was discovered after targeting Elastic's community Slack workspace. Multiple trojanized repositories were found with zero antivirus detections at the time of discovery, demonstrating the sophistication of this supply chain attack vector against software developers.
A sophisticated phishing campaign targeting Indian businesses and taxpayers leverages GST-related themes to distribute Remcos RAT through a multi-stage .NET framework. Threat actors impersonate Government of India GST communications using fraudulent refund notifications with convincing ARN references. The attack chain begins with a malicious RAR archive containing a .NET executable that employs bitmap-based payload concealment techniques. Through successive stages including Windows Health Optimizer Plus.dll and perfgurd.dll, the malware deploys Remcos RAT entirely in memory, establishing persistence via PowerShell scripts and registry modifications. Command-and-control infrastructure utilizes dynamic DNS services with randomized subdomains under aofmokighoig.hath.network. The deployed Remcos RAT enables remote command execution, keylogging, credential harvesting, file manipulation, and comprehensive system reconnaissance capabilities, representing a financially motivated cybercrime ope
In March 2026, threat actors leveraged AI-powered website builders to create typosquatting domains impersonating a Brazilian bank. The campaign employed ClickFix techniques, presenting victims with fake CAPTCHA and BSOD screens to trick them into executing malicious PowerShell commands. This delivered SmartRAT, a PowerShell-based banking trojan with capabilities including encrypted C2 communications, remote control of screen/keyboard/mouse, credential theft through keylogging and banking overlays, and QR code interception for transaction fraud. The malware establishes persistence via scheduled tasks and Windows services, and targets Brazilian financial institutions, payment platforms, and cryptocurrency exchanges. The threat actors' C2 panel contained critical authentication flaws allowing client-side bypass, suggesting deployment without adequate security review.