Summary
Attackers are hijacking trusted infrastructure at scale, with Gootloader leveraging over 1,000 compromised WordPress sites to funnel victims into ransomware schemes that have netted more than $28 million in payouts. Cyware spotlights how these footholds let criminals pivot deep into business systems, turning a single infection into a company-wide crisis.
A supply-chain breach at Optimizely injected malicious JavaScript into high-profile websites, exposing users to payment card skimming, cryptocurrency theft, and iOS exploits. Security researchers traced the campaign to 188 infected scripts, with attackers cashing out roughly $7,000 before cleanup efforts reduced the threat to a single compromised file.
A critical ScreenConnect vulnerability (CVE-2026-84869, CVSS 9.9) is being weaponized to seize MSP technician accounts and breach downstream customer networks in as little as nine seconds. The flaw’s active exploitation underscores the urgent need for patching and vigilant monitoring across managed service environments.
Operation ShadowCloud is blending command-and-control traffic into everyday cloud platforms like Microsoft SharePoint and GitHub, targeting government networks in Asia and Africa. The campaign’s use of advanced backdoors and obfuscation techniques highlights the evolving tactics threat actors employ to evade detection and compromise sensitive organizations.
Top Malware Reported in the Last 24 Hours
FBI links Gootloader to ransomware payouts
Gootloader is a malware framework that brokers initial access for follow-on attacks by luring victims through poisoned search results and compromised websites. The FBI says it has been used at scale via hijacked WordPress sites, giving criminals a foothold they can later turn into ransomware deployment and extortion. For organizations that get pulled in, the impact often goes beyond one infected machine, as attackers can pivot into business systems and pressure victims into paying to restore operations. As of October 2026, the FBI reported over 1,000 infected WordPress sites and 216 domains under its control, tied to over $28 million in ransom payments. The FBI’s advice includes contacting the bureau if compromised, blocking known malicious domains tied to the activity, and conducting regular security audits to spot unusual behavior on WordPress sites.
Optimizely supply-chain hack skims user data
A supply chain attack on Optimizely led to malicious JavaScript being served through customer projects, exposing visitors on high-profile websites to cryptocurrency theft, payment card skimming, and iOS exploit delivery. Security researcher Scott Helme found 188 infected Optimizely JavaScript files, and cleanup efforts later reduced that to one remaining compromised script. The malicious code also collected browser cookies, visited URLs, and screenshots, turning routine web browsing into an opportunity for account takeover and fraud. Cryptocurrency wallets linked to the activity received about $7,000 in transactions, offering a glimpse of the campaign’s monetization. For organizations using Optimizely, the recommended next steps cited in reporting include reviewing JavaScript files for unauthorized modifications and moving quickly on patching and security audits.
SeverVault sells infostealer logs with AI
SeverVault is a marketplace on DarkForums selling infostealer logs with AI-generated profiles that try to predict who an infected machine belongs to and what corporate access it might provide. It resells data from stealer families such as Vidar and Rhadamanthys, including session cookies, OAuth tokens, and other traces that can let criminals slip into accounts without needing a password. The AI-written summaries can be wrong, but they speed up targeting by highlighting devices that appear tied to government or financial-sector households, which listings reportedly price higher. For businesses, this kind of market turns a single employee infection into a broader risk of account hijacking and downstream intrusion when stolen sessions get replayed. The guidance alongside the report emphasizes limiting remote access to managed devices, treating shared devices as high-exposure endpoints, and monitoring for new exposures on dark web platforms.
Top Vulnerabilities Reported in Last 24 hours
Hackers hijack MSP access via ScreenConnect
Attackers are actively exploiting a critical ScreenConnect flaw to run malicious files on MSP technician accounts and pivot into customer environments (CVE-2026-84869, CVSS 9.9). Huntress says the bug allows a guest to trigger unauthorized file transfers and execute what gets pushed, turning an RMM session into a fast path to saved credentials, consoles, and downstream networks—one observed compromise reached full takeover in 9 seconds. Attackers are already exploiting this in the wild. Huntress reports the intrusions used batch files such as update_224ac089.bat alongside marker.txt, with Telegram bots used for reporting and coordination. A fix is available in ScreenConnect 26.6.5 or later; Huntress also called out interim steps including removing the TransferFiles permission, hunting for rogue ScreenConnect clients, and isolating compromised workstations.
Starlette bugs enable unauthenticated code execution
Two Starlette framework vulnerabilities allow remote code execution without authentication, putting Starlette-based services—and apps called out like LiteLLM—at risk (CVE-2026-42271, CVE-2026-48710). The issue stems from improper validation of the Host header, which attackers can manipulate to bypass access controls and reach endpoints that would normally be locked behind authentication. Attackers are already exploiting this in the wild, with activity reported as ongoing since July. SANS ISC highlighted exploitation patterns including crafted requests to endpoints such as /mcp-rest/test/tools/list, and provided an example showing a POST request where the attacker-controlled Host header is used to slip past protections. A fix is available in Starlette 1.0.1 or later, and SANS ISC advises building WAF rules around the underlying weakness—flagging unexpected characters in hostnames or allowing only specific hostnames—rather than chasing a single exploit pattern.
Firefox and GitLab ship major security fixes
Mozilla’s Firefox 160 update fixes 99 security vulnerabilities, including high-severity bugs that could corrupt browser memory and enable sandbox escapes that break the boundary between a web page and the underlying system. Mozilla’s advisory lists issues including CVE-2026-124853, CVE-2026-124855, and CVE-2026-124907, with many findings credited to AI-driven research efforts including OpenAI Codex Security and a researcher named kylebot. No active exploitation has been reported for these Firefox flaws. Separately, GitLab released fixes for 37 vulnerabilities spanning CI/CD pipelines, GitLab Pages, and MCP Server, including high-severity issues such as CVE-2026-105739 and CVE-2026-108153 (JavaScript execution via unsanitized input), CVE-2026-18241 (protected pipeline variables exposed via a race condition), and CVE-2026-97187 (cross-namespace Kubernetes Secrets read, reported by maxilagv via HackerOne). Fixes are available by updating to Firefox 160 and upgrading self-managed GitLab to 19.5.6, 19.4.8, or 19.3.9.
Top Threat Actors Reported in Last 24 hours
Operation ShadowCloud hides C2 in cloud
Operation ShadowCloud is a cyber intrusion campaign focused on government networks in Asia and Africa, blending into everyday cloud services to keep command-and-control traffic out of sight. After first surfacing in early 2026, they have used Microsoft SharePoint, GitHub, and Ably to relay instructions while deploying backdoors dubbed GitDoor and GraphDoor. Those tools are built to be hard to analyze, using encryption and obfuscation methods including control-flow flattening and indirect calls, and can pull down plug-ins on demand while injecting payloads into memory. GitDoor is described as supporting 26 commands covering capabilities such as file management, keystroke logging, and process injection, while GraphDoor relies on plug-ins such as kl.ts, pr.ts, and scr.ts for keylogging, process listing, and screenshot capture. The campaign’s early activity included an intrusion at an unspecified organization in Taiwan, underscoring how government staff can become the first casualties when cloud platforms are repurposed as attacker infrastructure. Researchers Saurabh Sharma and Noushin Shabab documented the activity and said defenders may respond by blocking a cited IP address and watching for MeshAgent/MeshCentral Agent artifacts.
SeverVault sells stealer logs with AI
SeverVault is a marketplace on DarkForums that packages infostealer logs with AI-generated “profiles,” turning raw stolen data into quick-hit targeting intel for criminal buyers. The listings emphasize supposed access value — including government and financial-sector households — which can accelerate follow-on fraud or intrusions when attackers spot session cookies, OAuth tokens, or traces of corporate access. The catch is that these AI-written profiles can be inaccurate, which can still cause real-world harm by pushing criminals to pursue the wrong people or by falsely elevating a victim as a high-value corporate doorway. The marketplace advertises logs tied to stealer families such as Vidar and Rhadamanthys, and prices rise when the data appears linked to sensitive roles. In practice, that means everyday credential theft can be rapidly “triaged” into targeted attacks against employers and agencies, even when the profiling is based on imperfect guesses.
Worldleaks ransomware hits Access Dental
Worldleaks is a ransomware actor that claimed responsibility for an attack on Access Dental Management LLC, a breach that later surfaced in notifications to US state authorities. They posted their claim on June 5, 2026, and the incident was disclosed to the Vermont Attorney General on November 12, 2026 and the Texas Attorney General on November 13, 2026. The reported impact in Texas alone was at least 341,039 people, a scale that can translate into years of identity and medical fraud risk for patients and staff. Exposed information included Social Security numbers, health records and medical information, health insurance information, names, addresses, and dates of birth. Affected individuals were notified by U.S. mail, reflecting the long tail of disruption that ransomware-driven data theft can impose even months after an initial claim.
Frequently Asked Questions
What is Gootloader? <b>Gootloader</b> is a malware framework that brokers initial access for follow-on attacks by luring victims through poisoned search results and compromised websites. The FBI says it has been used at scale via hijacked WordPress sites, giving criminals a foothold they can later turn into ransomware deployment and extortion.
What is Optimizely? A supply chain attack on <b>Optimizely</b> led to malicious JavaScript being served through customer projects, exposing visitors on high-profile websites to cryptocurrency theft, payment card skimming, and iOS exploit delivery. Security researcher <b>Scott Helme</b> found <b>188</b> infected Optimizely JavaScript files, and cleanup efforts later reduced that to <b>one</b> remaining compromised script.
What is SeverVault? <b>SeverVault</b> is a marketplace on DarkForums selling infostealer logs with AI-generated profiles that try to predict who an infected machine belongs to and what corporate access it might provide. It resells data from stealer families such as Vidar and Rhadamanthys, including session cookies, OAuth tokens, and other traces that can let criminals slip into accounts without needing a password.
What is CVE-2026-84869? Attackers are actively exploiting a critical ScreenConnect flaw to run malicious files on MSP technician accounts and pivot into customer environments (<b>CVE-2026-84869</b>, CVSS <b>9.9</b>). Huntress says the bug allows a guest to trigger unauthorized file transfers and execute what gets pushed, turning an RMM session into a fast path to saved credentials, consoles, and downstream networks—one observed compromise reached full takeover in <b>9 seconds</b>.
What is CVE-2026-42271? Two Starlette framework vulnerabilities allow remote code execution without authentication, putting Starlette-based services—and apps called out like LiteLLM—at risk (<b>CVE-2026-42271</b>, <b>CVE-2026-48710</b>). The issue stems from improper validation of the <i>Host</i> header, which attackers can manipulate to bypass access controls and reach endpoints that would normally be locked behind authentication.
What is Firefox? Mozilla’s Firefox <b>160</b> update fixes <b>99</b> security vulnerabilities, including high-severity bugs that could corrupt browser memory and enable sandbox escapes that break the boundary between a web page and the underlying system. Mozilla’s advisory lists issues including <b>CVE-2026-124853</b>, <b>CVE-2026-124855</b>, and <b>CVE-2026-124907</b>, with many findings credited to AI-driven research efforts including OpenAI Codex Security and a researcher named <i>kylebot</i>.
What is Operation ShadowCloud? <b>Operation ShadowCloud</b> is a cyber intrusion campaign focused on government networks in Asia and Africa, blending into everyday cloud services to keep command-and-control traffic out of sight. After first surfacing in early 2026, they have used Microsoft SharePoint, GitHub, and Ably to relay instructions while deploying backdoors dubbed <b>GitDoor</b> and <b>GraphDoor</b>.
What is SeverVault? <b>SeverVault</b> is a marketplace on DarkForums that packages infostealer logs with AI-generated “profiles,” turning raw stolen data into quick-hit targeting intel for criminal buyers. The listings emphasize supposed access value — including government and financial-sector households — which can accelerate follow-on fraud or intrusions when attackers spot session cookies, OAuth tokens, or traces of corporate access.
What is Worldleaks? <b>Worldleaks</b> is a ransomware actor that claimed responsibility for an attack on Access Dental Management LLC, a breach that later surfaced in notifications to US state authorities. They posted their claim on <b>June 5, 2026</b>, and the incident was disclosed to the <b>Vermont Attorney General</b> on <b>November 12, 2026</b> and the <b>Texas Attorney General</b> on <b>November 13, 2026</b>.


