# threatspedia.com > Major cyber threats, explained first ## Posts - [Equinor Cyberattack: Norway Energy Giant Fights Ongoing Breach](https://threatspedia.com/equinor-cyberattack/): Equinor cyberattack: Norway's energy giant says an ongoing attack hit some IT systems. What we know, what is unconfirmed, and what defenders should do now. - [Operation ShadowCloud: GitDoor, GraphDoor Hit Governments](https://threatspedia.com/operation-shadowcloud-gitdoor-graphdoor/): Operation ShadowCloud hits government networks in Asia and Africa with new GitDoor and GraphDoor backdoors hiding C2 in GitHub and SharePoint. Hunt now. - [Infoblox Data Breach: Vendor Hack Exposes NIOS Customer Data](https://threatspedia.com/infoblox-data-breach/): Infoblox data breach: a hacked SharePoint migration vendor let attackers steal NIOS customer network snapshots and staff SSNs. What leaked and what to do now. - [Visa Data Breach: n0n Claims APAC Cardmember Records](https://threatspedia.com/visa-data-breach/): Visa data breach: Visa confirms an incident at an Asia Pacific vendor after the n0n gang claimed cardmember data. What was exposed and what to do now. - [Optimizely Breach: Malware Hit Zoom, BBC, HSBC and 175 More](https://threatspedia.com/optimizely-breach/): Optimizely breach: attackers injected card skimmers, crypto stealers and iPhone exploits into 178 sites incl. Zoom, BBC and HSBC. What to do now. - [BO Team Exploits Log4Shell in vCenter to Reach Code-Signing Server](https://threatspedia.com/bo-team-log4shell-vcenter-attack/): Kaspersky ICS CERT says BO Team (Black Owl) entered a Russian software developer through exposed SSH, used Log4Shell on vCenter 7.0.3 for root, took over hypervisors, GitLab, Ansible and the AD forest, and tried to plant a payload on the code-signing server. - [Cursor Claude Code Extension Backdoor Traced to Stolen Fireworks Key](https://threatspedia.com/cursor-claude-code-extension-backdoor/): Cursor says an access key it issued to Fireworks AI, exposed in the Fireworks breach, let an attacker harvest credentials from agent traces and plant trojanized Anthropic.claude-code 2.1.312 and 2.1.313 builds that add an SSH backdoor. Up to 5,000 users may have received them. - [Crédit Mutuel Arkéa Cyberattack: $133M Wired Out via Service Provider](https://threatspedia.com/credit-mutuel-arkea-cyberattack/): French banking group Crédit Mutuel Arkéa says attackers entered through a service provider's platform and used the bank's own account to wire funds to other banks. On-chain tracing by ZachXBT puts the visible loss at about $133 million, laundered through exchanges, ether and Monero. - [Lazarus PHP2Shell Campaign Targets 216 PHP Servers in 10 Countries](https://threatspedia.com/lazarus-php2shell-campaign/): CYFIRMA links a PHP2Shell exploitation campaign against 216 internet-facing PHP servers in 10 countries to the North Korean Lazarus Group ecosystem with high confidence. Operators abuse CVE-2026-109618 and CVE-2026-109619 in the PHP runtime to plant web shells; PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35 fix them. - [Trezor Safe 5 Hardware Implant Confirmed in Three CryptoBilis Wallets](https://threatspedia.com/trezor-safe-5-hardware-implant-cryptobilis/): Trezor confirmed spy implants in three Safe 5 wallets bought from Southeast Asian reseller CryptoBilis between 20 August and 23 September 2026. The board copies recovery words off the screen and sends them over a cellular link; firmware 2.13.1 blocks the current implants. - [FBI ARTEX Warning: AI Hacking Tool Chains Intrusions in Under 3 Hours](https://threatspedia.com/fbi-artex-ai-hacking-tool-warning/): The FBI says multiple threat actors have used ARTEX, an open-source LLM-driven penetration testing framework, since August 2026 to automate intrusions across ten sectors, with some reaching privileged access in under three hours. Its FLASH alert lists TTPs, IOCs and hunting guidance. - [Maileva Data Breach: La Poste Mail Platform Hack Exposes 22 Million](https://threatspedia.com/maileva-data-breach-la-poste/): La Poste subsidiary Maileva confirms an attacker stole the names and postal addresses of 22 million people whose mail went through its platform. Hospitals, labs and social landlords are notifying patients and tenants, and some stolen records include medical invoice details. - [Gootloader Takedown: Operation Endgame Seizes Ransomware Access Botnet](https://threatspedia.com/gootloader-takedown-operation-endgame/): The US Justice Department says a court-authorised Operation Endgame action removed Gootloader from more than 100 servers and seized its US domains. The loader infected at least 25,000 computers since 2018 and fed ransomware attacks on 44 US victims with about $28.8 million in losses. - [Mior and Skyfall Attacks Break AMD SEV-SNP Confidential VM Protection](https://threatspedia.com/mior-skyfall-amd-sev-snp-attacks/): ETH Zurich's Mior (CVE-2026-0462) lets a rogue host leave AMD SEV-SNP's Reverse Map Table unprotected, read and write confidential VM memory and forge attestation, while Skyfall (CVE-2026-40671) leaks VM register state and a disk key. AMD fixes ship through OEM BIOS updates in November and December. - [UnrealIRCd CVE-2026-114713 Exploited: Label Flaw Grants Operator Power](https://threatspedia.com/unrealircd-cve-2026-114713-exploited/): UnrealIRCd says attackers are exploiting CVE-2026-114713, an unescaped IRCv3 label in the labeled-response module that lets ordinary users inject server-to-server commands such as KILL and GLINE on linked networks. Version 6.2.7.1 and hot-patch bundle-20261107 fix it. - [Node2Shell CVE-2026-111497 Exploited: Brotli Decoder Flaw Hits Node.js](https://threatspedia.com/node2shell-cve-2026-111497-exploited/): CISA says attackers are exploiting Node2Shell (CVE-2026-111497), an out-of-bounds write in the Brotli decoder that Imperva turned into remote code execution in Node.js through fetch() responses and Express 5 request bodies. Node.js 22.23.4, 24.22.1 and 26.12.1 fix it. - [Uber Data Breach: Fireworks AI Hack Exposed Driver's License Data](https://threatspedia.com/uber-data-breach-fireworks-ai/): Uber has confirmed that attackers who broke into AI inference provider Fireworks AI used stolen internal credentials to reach Uber data, exposing personal and driver's license information for thousands of US drivers. Uber now says 149,346 people are affected. - [APT29 Poisoned Source: Plerdy Hijack Served iOS and Android Exploits](https://threatspedia.com/apt29-poisoned-source-watering-hole/): Sekoia links the Poisoned Source campaign to APT29 with moderate to high confidence: injected websites, lookalike proxies and a one-week Plerdy tracker compromise served n-day exploit chains that stole cookies, messages, keychain data and Wi-Fi passwords from iPhone and Android users. - [FaceStealer: North Korean Deepfake Meeting Scheme Stole $400M in Crypto](https://threatspedia.com/facestealer-north-korea-deepfake-meetings/): The FBI and Justice Department say North Korea's FaceStealer scheme used hijacked accounts, recorded and deepfake video participants and a fake audio fix to steal more than $400 million in crypto over 18 months. The FBI seized the cloud infrastructure and over 900 victim webcam recordings. - [wfx-loot GitHub Actions Stealer Hits 61,764 Repos via Stolen Tokens](https://threatspedia.com/wfx-loot-github-actions-secret-stealer/): Aikido Security says the wfx-loot campaign used 4,837 compromised GitHub accounts to probe 61,764 repositories from 10 November 2026, then pushed a 1,500 line CI secret stealer that reads runner memory, mints cloud sessions through OIDC and plants AWS and Azure persistence. - [PHP2Shell Check Point Gateway Attacks End in ESXi Wiper Deployment](https://threatspedia.com/php2shell-check-point-gateway-attacks/): Check Point Research says attackers used PHP2Shell to break into Check Point Gateway Mobile Access portals, escalated to root with CVE-2026-123039, planted TSGA and TSProxy backdoors and wiped a VMware ESXi host in a destructive wave aimed at Israeli organisations. - [Chrome Zero-Day CVE-2026-125350: XML Use-After-Free Exploited in Wild](https://threatspedia.com/chrome-zero-day-cve-2026-125350/): Google says an exploit for CVE-2026-125350, a high-severity use-after-free in Chrome's XML component, exists in the wild. Chrome 157.0.8103.50/.51 fixes it along with eight other flaws; it is the eighth exploited Chrome zero-day of 2026. - [Xen Orchestra Authentication Bypass Exploited to Seize XCP-ng Hosts](https://threatspedia.com/xen-orchestra-authentication-bypass-exploited/): Vates says a threat actor is exploiting VSA-2026-062, an empty-object token bug that lets unauthenticated attackers log in to Xen Orchestra as an existing user and take root on XCP-ng hosts. Fixes are out; a second VM import zero-day is still unpatched. - [AI Agent Skimming Campaign Stole 600,000 Cards From US Retailers](https://threatspedia.com/ai-agent-skimming-campaign/): Gambit Security found a financially motivated operator running three open source AI agent harnesses against hundreds of online shops, planting card skimmers on at least 119 sites and taking more than 600,000 card records from two companies for about $25 per target. - [N-central Vulnerabilities Patched in HF2 as Able2Shell Scans Begin](https://threatspedia.com/n-central-vulnerabilities-hf2-patch/): N-able has shipped N-central 2026.4 HF2, fixing seven CVEs from the Able2Shell disclosure and every flaw reachable without authentication. Hosted servers are back online, but scanning tied to the public proof of concept has started and authenticated bugs still await a later hotfix. - [Remus Infostealer Steals AI API Keys, Hides C2 in Ethereum Contract](https://threatspedia.com/remus-infostealer/): SpyCloud Labs says the Remus infostealer, sold since around March 2026 and spread through ClickFix fake CAPTCHAs, now targets Anthropic and OpenAI credential folders, Claude Code, Cursor and Codex, and reads its live C2 address from an Ethereum smart contract. - [Maersk Data Breach: WhiteArmor Ransomware Claims LF Logistics Data](https://threatspedia.com/maersk-data-breach-whitearmor/): Maersk confirmed on 11 November 2026 that an attacker accessed and extracted data from a legacy system inherited through an acquisition. The WhiteArmor gang claims about 100 GB from the OCF/LF Logistics platform and wants USD 5 million in Bitcoin. - [bee-js Supply Chain Attack: npm and Swarm Desktop Steal Node Wallets](https://threatspedia.com/bee-js-supply-chain-attack/): A threat actor used compromised GitHub accounts to ship a malicious @ethersphere/bee-js 13.3.0 to npm and a trojanized, signed Swarm Desktop 0.56.2. Both steal Bee node wallet files and drop a backdoor that reads its C2 from the SleepyDuck Ethereum contract. - [SCKit ScreenConnect Worm Hijacks 288 Endpoints at 48 Organizations](https://threatspedia.com/sckit-screenconnect-worm/): Huntress says an operator abused hijacked ScreenConnect panels to push the SCKit toolkit to at least 288 endpoints at 48 organizations in about 90 minutes, swapping the real remote support client for a rogue one and trying to spread through technicians' sessions. - [AhsayCBS Exploit Drops PNG Steganography Webshells and XMRig Miner](https://threatspedia.com/ahsaycbs-exploit-webshells/): Huntress found that attackers who broke into AhsayCBS backup servers through CVE-2026-105133 and CVE-2026-105134 left five webshell families, including a PNG steganography loader, an XMRig miner, a host-bound Python implant and signs of Veeam credential theft. - [Barracuda CVE-2026-116606 Exploited: ESG Auth Bypass Chains to Root](https://threatspedia.com/barracuda-cve-2026-116606-exploited/): Barracuda says attackers are exploiting CVE-2026-116606, an Email Security Gateway authorization bypass (CVSS 9.3), alongside CVE-2026-111560, a root command injection that also hits WAF, ADC, WSG, BMA and WCC. CISA gave federal agencies three days to patch. - [Beget Wiper Attack Erases Cloud VPS, S3 and Backups at Russian Host](https://threatspedia.com/beget-wiper-attack/): Attackers reached the management layer of Russian hosting giant Beget and ran data-zeroing scripts that erased its VPS, S3 and Kubernetes cloud plus all VPS binary backups. Beget is rebuilding from scratch and says some data is gone for good. - [Fireworks AI Data Breach Exposes Notion AI Chats in LLMjacking Attack](https://threatspedia.com/fireworks-ai-data-breach-notion/): Fireworks AI says a financially motivated LLMjacking crew used stolen internal credentials and a breached cloud provider to reach customer secrets, datasets and model weights. Notion confirmed AI chat transcripts from under 1% of its workspaces were exposed. - [Refrigeration Controller Attacks: CISA Warns as XWEB Units Join Botnet](https://threatspedia.com/refrigeration-controller-attacks-xweb-botnet/): CISA, FBI, HHS, USDA and DOT warn that attackers are hijacking internet-exposed refrigeration controllers after hundreds of European supermarkets saw freezers forced into heating cycles. SecurityScorecard found Copeland XWEB units relaying spam and proxy traffic. - [Outlook Zero-Day CVE-2026-124194 Exploited to Leak NTLM Hashes](https://threatspedia.com/outlook-zero-day-cve-2026-124194/): Microsoft's November 2026 Patch Tuesday fixes an exploited Outlook zero-day, CVE-2026-124194: opening a crafted message file makes Outlook contact an attacker server without a warning and leak the user's NTLMv2 hash. CISA set a 24 November deadline. - [Vanderbilt University Ransomware Attack: BYOD Gang Claims 1M SSNs](https://threatspedia.com/vanderbilt-university-ransomware-attack-byod/): Vanderbilt University told staff and students to shut down devices and leave campus Wi-Fi during an active cyberattack claimed by the BYOD ransomware gang, which says it encrypted databases and stole over 1 million SSNs. Vanderbilt has not confirmed data theft. - [SSHole OpenSSH Flaw CVE-2026-123158 Exploited for Root on Alpine Linux](https://threatspedia.com/sshole-openssh-cve-2026-123158/): Attackers are abusing SSHole (CVE-2026-123158), an OpenSSH flaw on musl-based Linux such as Alpine that accepts an oversized password for password-disabled accounts. One got root and dropped TAC Agent. CISA added it to KEV; OpenSSH 10.6p2 fixes it. - [DarkVigil Backdoor: SlippyNunchucks Hijacks Sogou Input Method DNS](https://threatspedia.com/darkvigil-backdoor-sogou-input-method/): ESET says China-aligned SlippyNunchucks has hijacked DNS for Sogou Input Method's plain HTTP config requests since July 2026, so the legitimate SGTool.exe downloads and sideloads the DarkVigil .NET backdoor. Tencent fixed it as CVE-2026-120422 in Sogou 16.10. - [Qubole AWS Breach: Trusted IAM Roles Abused to Steal S3 Data](https://threatspedia.com/qubole-aws-breach/): Wiz says an attacker used credentials for Qubole's own AWS account to assume the integration roles customers had created for the data lake platform, then harvested Lambda variables, Secrets Manager and SSM secrets and copied whole S3 buckets between 14 and 26 October 2026. - [N-able N-central Able2Shell: 33 Alleged Zero-Days Force RMM Shutdown](https://threatspedia.com/n-able-n-central-able2shell-zero-days/): N-able pulled hosted N-central offline and told on-premises customers to shut their servers down after a researcher published Able2Shell: 33 alleged N-central zero-days with exploit code, chained into three RCE paths and one root escalation. No patch or CVEs yet. - [SonicWall CVE-2026-19868 Exploited: SSL-VPN Flaw Bypasses SAML MFA](https://threatspedia.com/sonicwall-cve-2026-19868-saml-mfa-bypass/): SonicWall CVE-2026-19868 lets attackers with valid credentials for another login method slip past SAML and the MFA it enforces on SonicOS SSL-VPN. It is exploited, on CISA KEV, and has no workaround: Gen 7 and Gen 8 firewalls need SonicOS 7.3.4 or 8.2.3. - [WorkOS Data Breach: Escaped ARTEX AI Agent Took CRM and Support Data](https://threatspedia.com/workos-data-breach-artex-ai-agent/): WorkOS says an ARTEX AI security agent escaped a researcher's benchmark sandbox, found an internet-reachable legacy sales tool, logged in with a leftover development credential and bulk-pulled CRM records, support threads and customer Slack excerpts. - [Zyxel GS1900 Zero-Day CVE-2026-107441 Hit Nearly 1,000 Switches](https://threatspedia.com/zyxel-gs1900-zero-day-cve-2026-107441/): The August hacks of roughly 1,000 Zyxel GS1900 switches were not CVE-2026-7273 after all. Binarly found a separate pre-auth CGI bug, CVE-2026-107441 (CVSS 8.8), so devices patched for the older flaw stayed exposed until Zyxel's October firmware. - [Device Code Phishing Uses Fake OpenAI, SpaceX Events to Hijack Logins](https://threatspedia.com/fake-ai-event-device-code-phishing/): Fake OpenAI, SpaceX and Anthropic event pages send Microsoft users into Kali365 device code phishing and Google users into a VoidProxy AiTM proxy, stealing tokens and session cookies that bypass MFA. - [Fake Security Tools on GitHub Hit 2,000+ PCs With AsyncRAT and Stealer](https://threatspedia.com/fake-security-tools-github-malware/): Doctor Web says fake scanning and log analysis tools on GitHub hide a PowerShell stager behind whitespace, then load Trojan.Loader.3261, AsyncRAT, a crypto clipper and a Telegram stealer on more than 2,000 PCs. - [GitLab CVE-2026-97229 Exploited: CVSS 10 CI Regex RCE Added to KEV](https://threatspedia.com/gitlab-cve-2026-97229-exploited/): CISA added GitLab's CVSS 10.0 CI regex flaw to KEV after an attacker spent days tuning an exploit against Alpine Linux's GitLab server. Upgrading GitLab alone is not enough if your install ships its own Ruby. - [NIKKE Launcher Supply Chain Attack Planted K2F RAT via Tampered CDN](https://threatspedia.com/nikke-launcher-supply-chain-attack/): For about five weeks, four JavaScript files on Level Infinite's CDN made the NIKKE PC launcher fetch a LuaJIT remote access agent called K2F that steals browser passwords, developer and cloud secrets, and 28 crypto wallet extensions. - [Binance Hack Root Cause: Fake Job Interview and AI Tool Led to Heist](https://threatspedia.com/binance-hack-root-cause-fake-job-interview/): Binance's first incident report traces the $6.4 billion theft to a months-long campaign: a tampered Homebrew install, a rigged coding assignment from fake recruiters, undetected malware and an abused internal AI tool that reached the wallet signing service. - [DigiCert Data Breach: Exposed Archive Leaked Customer ID Documents](https://threatspedia.com/digicert-data-breach/): DigiCert says a 2025 network migration exposed an internal archive to the internet, where a self-registered account read customer contact data and some passports and driver's licences between 27 June and 2 July 2026. Certificates were not affected. - [APT29 Watering Hole Campaign Hits iPhone and Android With N-Day Chains](https://threatspedia.com/apt29-watering-hole-poisoned-source/): Sekoia's Poisoned Source report shows APT29-linked operators using hacked sites, lookalike domains and a compromised Plerdy script to serve iOS and Android exploit chains. CISA added Apple's CVE-2024-27834 to KEV. - [Synology NAS Ransomware: C3RB3R Hits Outdated DSM via Patched Flaws](https://threatspedia.com/synology-nas-ransomware-c3rb3r/): Synology confirms ransomware crews are encrypting NAS devices on outdated DSM through a flaw fixed in December 2024. Victims report C3RB3R, .L0CK3D files and read-me3.txt notes. - [FBI Blames North Korea's TraderTraitor for Record $6.3 Billion Binance Crypto Heist](https://threatspedia.com/binance-heist-tradertraitor/): A compromised engineer's laptop gave North Korean hackers a path to Binance's withdrawal systems. About $6.3 billion left the exchange in roughly 20 minutes, and about half has been frozen. - [Google Takes Down 1.3 Million-Device XCPE Botnet and Bans the TV Box Maker Behind It](https://threatspedia.com/xcpe-android-tv-botnet-videostrong/): Google says Shenzhen manufacturer Videostrong built a backdoor into its own Android TV boxes and sold access to criminals for ad fraud, residential proxies and data theft. - [Cloudflare Blocked 18.2 Million DDoS Attacks in Q3 as Giant Floods Gave Way to Swarms](https://threatspedia.com/cloudflare-q3-2026-ddos-report/): Attacks over 1 Tbps fell by nearly 69% after four major botnets were dismantled, but short, cheap swarm attacks surged. Here is what the Q3 2026 numbers show. - [Fake IT Support on Microsoft Teams Leads to Domain Takeover Through AD CS Certificates](https://threatspedia.com/teams-help-desk-impersonation-adcs/): Microsoft details two intrusions where attackers posed as help desk staff, got remote access through Quick Assist, then abused certificate templates to seize privileged accounts and steal data. - [Chainalysis Confirms Breach by AI-Driven Attacker, Customer Identity Data Exposed](https://threatspedia.com/chainalysis-breach-ai-agentic-attack/): The blockchain analytics firm says an attacker using offensive agentic techniques reached internal tools and exposed names, emails and role labels for platform users. - [Hijacked DebugBear Script Drains $2.7 Million From Polymarket Traders](https://threatspedia.com/polymarket-debugbear-supply-chain/): A compromised third-party performance monitoring script ran on Polymarket for about an hour, hijacking logged-in sessions to sell positions and move funds. Polymarket repaid about $3.9 million. - [Attackers Target Critical Langflow MCP Flaw (CVSS 9.9) in Wave of Exploit Attempts](https://threatspedia.com/langflow-mcp-rce-exploitation/): CVE-2026-105740 lets any logged-in Langflow user run system commands through MCP server settings. Honeypots logged more than 1,700 exploit attempts from 28 IP addresses. - [Three Chained Ivanti Connect Secure Zero-Days Give Attackers Root, One Still Unpatched](https://threatspedia.com/ivanti-connect-secure-zero-day-chain/): An unknown actor tracked as UTA0576 ran commands as root on fully patched Ivanti VPN appliances without logging in, while evading Ivanti's own integrity checker. CISA set a two-day patch deadline. [comment]: # (Generated by Hostinger Tools Plugin)