Koffei

Daily CyberNews · 13/11/2026

‹ 12/11/202616/11/2026 ›
Today in 30 seconds
  • 🔴 Patch PHP2Shell and apply the latest Check Point Jumbo Hotfix: gateways were breached and ESXi hosts wiped.
  • 🔴 Update UnrealIRCd to 6.2.7.1: CVE-2026-114713 is exploited and in KEV.
  • 🔴 N-central Hotfix 2 is not enough: upgrade agents and probes and hunt for XMRig dropped via CVE-2026-90436.
  • 🔴 On-premises ScreenConnect customers: remove the RMM+ extension and check technician workstations.
  • 🟠 Upgrade NetScaler ADC, Gateway and Console: 12 flaws fixed, including an unauthenticated Console auth bypass.

Vulnerabilities 11

🔴 Exploited or active

PHP2Shell used to break into Check Point gateways, then root and wipe ESXi hosts

CVE-2026-109618CVE-2026-109619in KEVCheck Point

WhatAttackers used PHP2Shell on unpatched Check Point gateways, got root via CVE-2026-123039, and in one wave wiped VMware ESXi hosts at Israeli organisations.

DoPatch PHP2Shell, apply the latest Check Point Jumbo Hotfix per sk1000362, and hunt for zz*.tpl webshells.

Earlier coverage: 30/10
Follow-ups: 16/1117/1119/11
Details and sources
Check Point Research says attackers hit unpatched Security Gateways with the Mobile Access portal through the PHP2Shell pair CVE-2026-109618 (CVSS v4 9.2 / v3.1 9.8) and CVE-2026-109619 (CVSS v4 9.2), unauthenticated PHP RCE flaws fixed in PHP 8.2.35/8.3.36/8.4.27/8.5.12 and in CISA KEV since 26 October. They planted zz*.tpl webshells, then used CVE-2026-123039 (CVSS 7.8, local only: the root DynamicCLI daemon listens on a world-writable socket; affects Gaia R81.20 up to JHF Take 173, R82 up to Take 130, R82.10 up to Take 50, R82.20 up to Take 10; advisory sk1000362) to get root. A first wave from 30 October hit Israeli organisations and ended with the destroy_v6 wiper on VMware ESXi leaving a "DREAM_ANNEXATION.txt" note; a second wave from 2 November reached several countries without destruction. The TSGA and TSProxy implants are modified PhiliKit, a toolkit ESET ties to China-nexus UNC5221, but Check Point keeps attribution at low confidence (possible access broker or false flag). Patch PHP2Shell and apply the latest Jumbo Hotfix. Check Point (source) · SecurityWeek (source)
🟠 High impact

Two critical Red Hat Quay flaws let attackers read or overwrite any private image

CVE-2026-123710CVE-2026-110697CVSS 9.9Red Hat Quay

WhatTwo critical Quay flaws let attackers add a non-expiring signing key to pull or push any image, or list other tenants' robot tokens.

DoUpgrade Quay to a fixed release such as 3.18.3; until then block /keys at the proxy and audit for non-expiring key rotations.

Details and sources
CVE-2026-123710 (CVSS 9.8, no authentication) lets an anonymous token reach the key-rotation endpoint, so an attacker can add a non-expiring signing key and then pull or push images in any repository. CVE-2026-110697 (CVSS 9.9) abuses SQL LIKE wildcards (underscores in a namespace name) to list other tenants' robot accounts with their decrypted tokens. All Quay 3 and Quay 5 deployments are affected, including the Quay Operator and the OpenShift mirror registry; fixes ship in 3.9.28, 3.10.28, 3.12.24, 3.14.10, 3.15.9, 3.16.7, 3.17.7 and 3.18.3, and Quay.io is already fixed. Not exploited, not in KEV. Until patched, block /keys at the proxy and review audit logs for key rotations with no expiry. The Hacker News (source) · Red Hat (source)
🟠 High impact

Citrix patches 12 NetScaler and Secure Access Client flaws

Citrix NetScalerCVE-2026-120869CVE-2026-107412CVSS 9.5

WhatCitrix fixed NetScaler ADC and Gateway memory flaws that crash appliances and an unauthenticated auth bypass in NetScaler Console; no exploits known.

DoUpgrade NetScaler ADC and Gateway to 14.1-73.57 or 13.1-64.39, and NetScaler Console to 14.1-74.34.

Details and sources
bulletin CTX697305 fixes memory flaws in NetScaler ADC and Gateway that crash the appliance (DoS): CVE-2026-120868 (packet processing, every deployment), CVE-2026-107407 (SMPP), CVE-2026-107408 (DNS/DoH), CVE-2026-107409 (DoH), CVE-2026-107410 (Push), CVE-2026-107415 (Email OTP), CVE-2026-120864 (ICA proxy), all CVSS v4 9.3, and CVE-2026-107412 (Gateway/AAA virtual server, 9.5). Fixed in 14.1-73.57, 13.1-64.39, 14.1-FIPS 14.1-73.58 and 13.1-FIPS/NDcPP 13.1-37.293. CTX697320 fixes an unauthenticated auth bypass in NetScaler Console, CVE-2026-120869 (CVSS 9.3 in the CVE record; the article says 9.4), fixed in 14.1-74.34. The article also lists Secure Access Client local privilege escalations CVE-2026-118268 (Windows, to SYSTEM, fixed 26.11.1.6) and CVE-2026-118269 (Linux, to root, fixed 26.10.4), not yet published in the CVE list, and an Endpoint Analysis flaw under an ID (CVE-2026-111344) that actually belongs to an unrelated APITable bug, so treat those three as unconfirmed. Citrix is "not aware of any exploits". The Hacker News (source)
🟠 High impact

"SignDrop": unpatched Windows CLFS zero-day PoC gives SYSTEM

Windows CLFSSignDropzero-dayno patch

WhatA public, unpatched exploit abuses the Windows CLFS driver and Defender's update interface to gain SYSTEM on Windows 11 and Server 2025.

Details and sources
a researcher known as "MSNightmare" published a working exploit on Patch Tuesday that abuses how the CLFS driver (clfs.sys) creates its companion .cnpf file, redirecting it with symbolic links, and uses Microsoft Defender's update interface as the trigger (Defender must be on). It works where the CLFS Authentication "Mode" registry value is 0 or 1 and was tested on Windows 11 26H2 and Windows Server 2025 builds; an independent tester confirmed it, though not reliably. No CVE, no patch, no in-the-wild use reported. The Hacker News (source) · BleepingComputer (source)
🔴 Exploited or active

N-central exploit chain now used to drop miners on 21 servers, 19 of them already on Hotfix 2

N-able N-centralCVE-2026-90436XMRig

WhatThe 4EverAble2Shell chain dropped the XMRig miner on 21 N-central servers, 19 of them already on Hotfix 2.

DoUpgrade agents and probes to 2026.4 HF1 or later and hunt for a jup.sh script fetched from 157.20.241[.]16.

Earlier coverage: 12/1111/1110/11
Follow-ups: 18/1119/11
Details and sources
Huntback found the XMRig Monero miner, fetched by a jup.sh script from 157.20.241[.]16, on 21 N-central servers, mostly in the US, South Africa, Canada, the UK, Germany, France and Sweden; "Applying Hotfix 2 is not enough on its own". The 4EverAble2Shell chain pairs end-of-life Apache Axis 1.x CVE-2026-90436 (CVSS 9.8, SOAP fault class instantiation, pre-auth RCE, no fix planned: migrate to Axis2) with an N-central privilege escalation that has no CVE yet. N-able advises upgrading agents and probes to 2026.4 HF1 or later. CISA still rated the Axis flaw "no known exploitation" on 10 November and it is not in KEV. The Hacker News (source)
🟡 Worth knowing

Microsoft's automated audit yields 59 Linux kernel bug reports and 33 CVEs

MicrosoftLinux kernel33 CVEs

WhatMicrosoft's AI-assisted Linux kernel review produced 59 bug reports and 33 CVEs, with working root exploits built for eight; no in-the-wild use.

Details and sources
Microsoft says its AI-assisted kernel review produced 59 reports, 33 of them now CVEs, some also reachable on Android with SELinux enforcing; working root exploits were built for 8, with no in-the-wild use. Microsoft (source) · The Hacker News (source)
🟡 Worth knowing

CISA: six ICS advisories, no new KEV entries seen

CISAICS advisoriesSiemensno new KEV

WhatCISA released six ICS advisories, including Siemens building automation products, and added no KEV entries on Nov 12 to 13.

DoFederal agencies must remediate the two Barracuda ESG KEV entries by today.

Details and sources
CISA released six ICS advisories on 12 November, including ICSA-26-316-01 (Siemens Building Automation and Energy products) and ICSA-26-316-02 (Siemens Desigo, Gridscale). The KEV mirror shows no additions dated 12 or 13 November at the time of writing (cisa.gov itself could not be reached for this run). Today is the federal deadline for the two Barracuda ESG entries added on 10 November. CISA (source) · CISA (source)
🟠 High impact

Second Chrome update this week: 83 security fixes, five critical

Added after publishingGoogle Chrome83 fixesfive critical

WhatChrome 157.0.8103.78/.79 fixes 83 security bugs, including five critical use-after-free flaws that can lead to code execution.

DoUpdate Chrome to 157.0.8103.78/.79 and relaunch the browser.

Details and sources
Google shipped Chrome 157.0.8103.78/.79 for Windows and Mac and 157.0.8103.78 for Linux, two days after the previous update, with 83 security fixes; all five rated Critical are use-after-free bugs, which can lead to code execution. Malwarebytes lists no CVE IDs and does not say whether any is exploited; Google's own release notes were not read for this entry. Update and relaunch the browser. Malwarebytes (source)
🔴 Exploited or active

UnrealIRCd label-injection flaw exploited, in CISA KEV with a 16/11 deadline

Added after publishingCVE-2026-114713CVSS 8.6UnrealIRCdin KEV

WhatAn UnrealIRCd label-injection flaw lets any user on linked networks inject server commands, and it is exploited in the wild.

DoUpdate UnrealIRCd to 6.2.7.1; federal agencies must patch by Nov 16.

Follow-ups: 16/11
Details and sources
CVE-2026-114713 (CVSS 8.6, CWE-140) in UnrealIRCd's labeled-response module does not escape the IRCv3 "label" tag, so any user on a network of two or more linked servers can inject server-to-server commands that run with server or IRCOp authority. Affects 6.0.0 up to 6.2.7.1, fixed in 6.2.7.1 (hotfix 7 November). Exploitation began about a day after the hotfix ("we are seeing bad guys exploiting this issue on the internet right now", maintainer Bram Matthys, who calls it the most critical UnrealIRCd issue in 15 years); CISA added it to KEV on 13 November, with federal agencies due to patch by 16 November. Attackers and scale not stated. The Hacker News (source) · CISA (source)
🟠 High impact

PostgreSQL patches 52 flaws, 32 of them rated 8.8

Added after publishingCVE-2026-6460CVSS 8.8PostgreSQL52 flaws

WhatPostgreSQL fixed 52 flaws, including a type confusion letting object creators run code as the OS user and a pre-auth restart bug.

DoUpgrade to PostgreSQL 18.7, 17.12, 16.16, 15.20 or 14.25, and rebuild C extensions when upgrading.

Details and sources
the 12 November releases 18.7, 17.12, 16.16, 15.20 and 14.25 (14 is now end of life) fix 52 flaws. CVE-2026-6460 (CVSS 8.8, CWE-843 type confusion in ALTER ... SET DATA TYPE) lets a user who can create objects run code as the PostgreSQL OS user; CVE-2026-116322 (CVSS 5.3) lets an unauthenticated client restart PostgreSQL 18 servers through a pre-auth spinlock leak. pg_restore and psql also get fixes against crafted archives and compromised source servers. No exploitation reported (CISA SSVC: none); researcher Mehmet Ince published a crash-and-memory-leak PoC for 18.6 without the final code-execution step. Rebuild C extensions when upgrading. The Hacker News (source)
🟡 Worth knowing

Mior and Skyfall let a rogue host read AMD SEV-SNP confidential VMs

Added after publishingAMD SEV-SNPCVE-2026-0462CVE-2026-40671MiorSkyfall

WhatMior and Skyfall show a rogue hypervisor can break integrity or leak data of AMD SEV-SNP confidential VMs; PoCs are public, no exploitation.

DoGet BIOS updates with the AMD platform firmware from your server maker when released.

Details and sources
ETH Zurich (Mior) and ETH Zurich with Google (Skyfall) show hypervisor-level attacks on AMD EPYC 7003, 8004, 8005, 9004 and 9005 confidential VMs. Mior, CVE-2026-0462 (CVSS 4.0 5.9, AMD-SB-3059), redirects MMIO to break guest memory integrity (the researchers say it can also read memory and forge attestation reports) and needs BIOS and PCIe-device control; Skyfall, CVE-2026-40671 (CVSS 4.0 4.9, AMD-SB-3061), leaks guest register data through VMRUN error handling and builds on the Transient Scheduler Attacks. A related EDK II SMRAM leak is CVE-2026-18448 (CVSS 4.0 5.6). AMD's platform firmware is due in November (Skyfall) and December (Mior), embedded parts in 2027; get BIOS updates from your server maker. PoCs are public; no exploitation reported. The two AMD CVE records are not yet published, so their values come from the AMD bulletins. The Hacker News (source)

Breaches / Incidents 5

🟠 High impact

TSMC confirms current and former employee data leaked via a cloud provider

TSMCemployee datacloud provider

WhatTSMC confirmed current and former employee data leaked via an unnamed cloud provider; sellers claim 46,000 records, which is unverified.

Details and sources
in a Form 6-K signed 12 November, TSMC says some employee data "was made available on an external website without authorization" and matches data held by an unnamed cloud service provider; it sees no compromise of its own systems and no customer or operational data. Forum sellers reportedly offered 46,000 records (names, national ID numbers, bank details) for 3 BTC; those figures are unverified and TSMC gave no count. TSMC (source) · Cyber Security News (source)
🔴 Exploited or active

ConnectWise narrows the ScreenConnect intrusions to the "RMM+" extension

ConnectWise ScreenConnectRMM+ extensionCVE-2026-84869

WhatConnectWise ties the ScreenConnect intrusions to the third-party RMM+ extension, present in about 400 partner environments, and disabled it in cloud instances.

DoOn-premises customers must remove RMM+ and check technician workstations, since credential resets alone may not evict attackers.

Earlier coverage: 12/11
Follow-ups: 16/1118/11
Details and sources
ConnectWise says its investigation points to the third-party RMM+ extension, found in about 400 partner environments (presence alone is not proof of compromise). It disabled RMM+ in cloud instances; on-premises customers must remove it themselves, and technician workstations may already be compromised, so credential resets alone may not evict attackers. Huntress sees no new flaw and suspects stolen credentials or hijacked sessions, while Blackpoint ties the toolkit to CVE-2026-84869 (CVSS 9.9, client-side guest-to-host file execution, fixed in 26.6.5, in KEV since 11 September). The Record (source)
🟡 Worth knowing

Maersk says the incident is limited to a legacy system from an acquisition

MaerskWhiteArmorlegacy system

WhatMaersk says the incident is limited to one legacy system from an acquisition, while WhiteArmor's 100GB claim remains unverified.

Earlier coverage: 12/11
Details and sources
Maersk's statement says one local legacy system was hit, with no impact on other Maersk, customer or partner systems (including API/EDI links) and findings that "suggest no sensitive data has been disclosed". WhiteArmor claims it hit LF Logistics (acquired in 2022), took 100GB and wants $5 million within 30 days; the claim is unverified. The Record (source)
🟡 Worth knowing

London hotel left its Booking.com admin login on a whiteboard

London hotelBooking.comexposed credentials

WhatA London hotel left its Booking.com admin login on a whiteboard behind the front desk; no misuse was reported.

Details and sources
a researcher spotted the username and password behind the front desk; anyone with them could change or cancel bookings, see guest details and message guests as the hotel. No misuse was reported and the hotel changed its practices. The Register (source)
🔴 Exploited or active

Trezor confirms hidden implants in Safe 5 wallets sold by one reseller

Added after publishingTrezor Safe 5hardware implantCryptoBilis

WhatThree Trezor Safe 5 wallets sold by reseller CryptoBilis carried a hidden board that copies the recovery phrase and sends it to attackers.

DoCryptoBilis buyers since March should move funds to a new wallet with a new recovery phrase; firmware 2.13.1 blocks known implants.

Earlier coverage: 02/11
Details and sources
Trezor confirmed three Safe 5 wallets bought from reseller CryptoBilis between 20 August and 23 September carry an extra circuit board that copies the recovery phrase (and any on-screen passphrase) during backup or restore and sends it to attacker servers over the mobile network; firmware and Secure Element are untouched. No funds reported lost; no implants found in other resellers' stock or other models, and who fitted them is not known. Firmware 2.13.1 blocks the known implants but cannot protect a phrase already captured: CryptoBilis buyers since March should move funds to a new wallet with a new recovery phrase. The Hacker News (source)

Malware / Ransomware 5

🔴 Exploited or active

"TypoPrompting": 186 fake AI domains push ClickFix lures and browser malware

TypoPromptingClickFixfake AI domains

What186 lookalike domains imitating 27 AI brands push ClickFix lures, a malicious MSIX app and search-hijacking Chrome extensions with about 972,000 users.

Details and sources
Island found 186 lookalike domains imitating 27 AI brands (Claude, ChatGPT, Gemini, Copilot, Perplexity, Cursor and others). A fingerprinting gate sends visitors to ClickFix fake Cloudflare checks (PowerShell or macOS shell commands), to the "Net Guard" MSIX app that runs attacker JavaScript, swaps the search engine and installs Chrome extensions, or to search-hijacking extensions such as Quick Search (16 Chrome Web Store listings, about 972,000 users). The same gate fronts 211 domains imitating banks and insurers. The Hacker News (source)
🟠 High impact

GSocket abused for stealthy C2 on hacked Linux web servers

GSocketLinux web serversC2

WhatAttackers install GSocket on hacked Linux web servers for C2 over a public relay network, with detections up 356% from March to August.

Details and sources
Microsoft says attackers install the open-source GSocket tool after exploiting web apps; its traffic rides a public relay network, leaving no attacker IP or domain to block. Detections rose 356% between March and August 2026, 87% of attributable cases on PHP/cPanel/Apache hosting, with education and tech/telecom most affected. Microsoft (source)
🟠 High impact

Grandoreiro returns with date-seeded domains against Latin America

GrandoreiroLatin Americaphishing

WhatSpanish-language phishing in five Latin American countries delivers a VBScript and Delphi loader that fetches the Grandoreiro banking malware.

Details and sources
Arctic Wolf reports Spanish-language phishing in Argentina, Chile, Colombia, Mexico and Peru serving a ZIP with a VBScript that starts a Delphi loader disguised as a PDF reader, which then fetches Grandoreiro. Arctic Wolf (source) · The Hacker News (source)
🟠 High impact

RevStealer spreads through fake AI apps and hijacked YouTube channels

RevStealerfake AI appsYouTube

WhatRevStealer spreads through 100+ domains, game cheats and fake Claude apps, stealing browser data, passwords, crypto wallets and gaming accounts.

Details and sources
RevStealer is pushed via 100+ domains, game cheats and trojanized Electron apps posing as Claude, uses direct syscalls to dodge hooks and a Polygon blockchain dead-drop as fallback C2, and targets browsers, password managers, crypto wallets and gaming accounts. Akatsuki Legion (source) · The Hacker News (source)
🟡 Worth knowing

GTA VI scams in full swing before the game's release

Added after publishingGTA VI scamsVidarnjRATDCRat

WhatAhead of GTA VI's release, scammers run phishing pages and fake installers delivering Vidar, njRAT, DCRat and the Chaos wiper.

Details and sources
ESET lists phishing pages imitating Rockstar, PlayStation, Xbox and Steam, fake beta or early-access offers paid in crypto ($40 to $250), cloned stores and fake "leaks". A 1.1 MB gta6_installer.exe carried the Vidar infostealer; 100 GB+ fake ISOs delivered njRAT, DCRat, Mercurial Grabber and the Chaos wiper, which overwrites files over 200 MB; Netcraft counted 121 "CyberLeek" domains, 100 of them registered in August. ESET (source)

Threat Actor Activity 6

🟠 High impact

Silent Lynx used an AI agent to raid Tajik railway and government systems

Silent LynxYoroTrooperTajikistan

WhatSilent Lynx used an AI agent to steal Defense Ministry and other government data via Tajikistan's national railway network from August to October.

Details and sources
CloudSEK found an exposed Silent Lynx (aka YoroTrooper, Cavalry Werewolf) server with over a dozen GB of stolen data. From mid-August to early October the group sat in Tajikistan's national railway network, used one employee's credentials to reach the national e-document platform and took Defense Ministry, National Guard and State Committee for National Security material plus data on 100+ government bodies; an Uzbek Smart Forestry server was also hit and Turkmenistan lures were prepared. The operator drove the open-source Hermes agent on DeepSeek models through a Russian-language Telegram bot. CloudSEK sees a Kazakhstan nexus at medium confidence, with no proven state link. The Record (source) · CloudSEK (source)
🟠 High impact

China-linked CL-STA-1238 runs a forked AI pentest framework against chipmakers and defense

CL-STA-1238Cairn forksemiconductorsdefense

WhatA China-linked cluster used a forked AI pentest framework across 25 operations against at least 19 organisations, confirming access at 10.

Details and sources
Unit 42 says the cluster used a customised fork of the open-source Cairn framework with GLM and DeepSeek models across 25 operations, targeting at least 19 semiconductor, defense and civil-society organisations and confirming access at 10. Unit 42 (source)
🟡 Worth knowing

"BluePhishing" abuses AI-agent blueprint consent in Microsoft Entra

BluePhishingMicrosoft Entraconsent phishing

WhatOne-time consent links trick Entra users into approving a malicious AI-agent blueprint, giving attackers agent identities and tokens in the tenant.

Details and sources
Proofpoint describes one-time consent links that get Entra users to approve a malicious third-party agent blueprint, handing the attacker agent identities and tokens inside the victim tenant. Proofpoint (source) · The Hacker News (source)
🟠 High impact

SilentAir hijacks Asian telecom signalling to spy on Taiwanese mobile users

Added after publishingSilentAirSS7DiameterTaiwan

WhatA state-level group abused telecom signalling in six Asian countries to divert Taiwanese users' calls and texts, including some 2FA codes.

Details and sources
Enea named SilentAir, a group it assesses with high confidence as state-level, that sent SS7 and Diameter messages into the IPX interconnect from compromised operator equipment in at least six Asian countries. Forged UpdateLocation messages diverted calls and texts of Taiwanese mobile users, including political and government VIPs, and some targets' 2FA codes were intercepted right after diversion (no account takeover confirmed); bursts of night-time silent SMS likely checked that diversions held. Activity has been tracked since early 2025 and is disrupted for now. Enea links it to China only with low confidence (Taiwan focus, UTC+8 hours, pauses on Chinese holidays); a tie to Liminal Panda (SIGweaver / SCTPE, GTPDoor) is possible but unconfirmed, and SilentAir showed better opsec. The Hacker News (source)
🟠 High impact

Amazon: Chinese-nexus groups run cybercrime campaigns with ARTEX AI agents

Added after publishingARTEXCVE-2025-3248CVE-2026-28797AI credentials

WhatAmazon saw Chinese-nexus groups run ARTEX AI agents to steal AI-model credentials from gateways and bypass crypto exchange identity checks with deepfakes.

Earlier coverage: 11/1109/1106/11
Details and sources
Amazon's MadPot sensors captured, from 3 October to 5 November, AI-agent requests carrying the agents' own campaign records: mass theft of AI-model credentials from gateways and app builders (One API, LiteLLM, Dify, Langflow) with CVE-2025-3248 (Langflow, CVSS 9.8, in KEV since May 2025) and CVE-2026-28797 (RAGFlow SSTI, CVSS 8.8, no patch listed) in the playbook, plus deepfake bypass of identity checks at two crypto exchanges. Amazon assesses with moderate confidence that disparate Chinese-nexus groups ran ARTEX (planner, worker, verification and compression agents sharing a knowledge graph); no group is named and success against each CVE is not confirmed. Amazon (source)
🔴 Exploited or active

FBI details BlueNoroff's FaceStealer deepfake-meeting scheme

Added after publishingBlueNoroffAPT38FaceStealerdeepfakes

WhatThe FBI says BlueNoroff stole over $400M in 18 months via fake Zoom or Teams meetings with deepfakes leading to ClickFix-style malware installs.

Details and sources
the FBI says BlueNoroff (APT38, UNC1069, Sapphire Sleet) stole more than $400M in 18 months by luring victims into fake Zoom or Teams meetings from hijacked accounts, with prerecorded videos and deepfakes, then a fake audio error that leads to a ClickFix-style malware install; each victim is used to reach their contacts. The scheme overlaps Kaspersky's GhostCall (October 2025) and targeted the maintainer of the Axios npm package. Arctic Wolf counts 677 targets in 58 countries, over 80% founders or C-suite in crypto and finance; the actors now scrape YouTube interviews and talks to build deepfakes of specific people. The Hacker News (source)

Law Enforcement 4

🟡 Worth knowing

Operation Endgame dismantles Gootloader, CastleRAT and StealC infrastructure

Operation EndgameGootloaderCastleRATStealC

WhatPolice seized or blocked 257 domains and 235 servers of Gootloader, CastleRAT and StealC and recovered about 6.4 million victim datasets.

DoCheck affected credentials through HaveIBeenPwned or Shadowserver, where the recovered data is being shared.

Details and sources
Europol and Eurojust coordinated police in Bulgaria, Canada, Denmark, Germany, Moldova, the Netherlands, Poland, the UK and the US to seize or block 257 domains and 235 servers, recovering about 6.4 million victim datasets (6M+ credentials) now shared through HaveIBeenPwned and Shadowserver. The DOJ says the FBI, under court order, removed Gootloader from 100+ servers (many hacked WordPress sites); it has infected at least 25,000 computers since 2018 and is tied to ransomware against 44+ US victims with about $28.8M in losses. StealC and CastleRAT servers were also taken down. No arrests announced. BleepingComputer (source) · DOJ (source) · Europol (source)
🟡 Worth knowing

Operation Endgame: Germany and the Netherlands add victim and infrastructure figures

Added after publishingOperation EndgameBKAGootloaderCastleRAT

WhatGermany and the Netherlands say the seized logins belong to over 40,000 victims and detail further Gootloader and CastleRAT takedowns; no arrests.

DoUse the Dutch police Check your hack service to check an email address against the CastleRAT data.

Details and sources
Germany's BKA says the ~6.4 million seized logins belong to more than 40,000 victims. For Gootloader, police took down 14 servers and 218 domains and Denmark's NSK cleaned 681 hacked websites; German police put its infections in the millions since 2020, far above the DOJ's 25,000. 120 servers and 39 domains of CastleRAT were seized, and StealC's operators had rebuilt after a June operation, partly on the Virtualine/Omegatech bulletproof host shut down in late October. Still no arrests: Dutch police continue investigating Gootloader's main suspect, and their "Check your hack" service lets people check an email address against the CastleRAT data. The Hacker News (source)
🟡 Worth knowing

DOJ moves to seize $1.2B of the Binance theft and takes down North Korea's "FaceStealer" platform

TraderTraitorBlueNoroffBinanceDOJ

WhatUS courts authorised seizure of over $1.2B of the Binance theft, and the FBI seized FaceStealer's cloud infrastructure.

Earlier coverage: 05/1104/1103/11
Details and sources
about $2.7B of the funds TraderTraitor stole from Binance has been frozen, and courts authorised seizure of more than $1.2B of it. The FBI also seized the cloud infrastructure behind FaceStealer, a fake Zoom/Teams meeting scheme tied to APT38/BlueNoroff that held 900+ secretly recorded victim webcam videos and drained over $400M in 18 months. Elliptic separately traced the laundering through 20,000+ wallets on 16 blockchains, with about $3.0B still in linked addresses. CyberScoop (source) · The Record (source) · Elliptic (source)
🟡 Worth knowing

Police investigate bomb-hoax emails to European schools sent from hijacked accounts

Bomb hoax emailsEuropean schoolshijacked accounts

WhatPolice investigate bomb-hoax emails sent to thousands of European schools from hijacked accounts; nothing is attributed.

Details and sources
Police Scotland, the Garda and the PSNI are assessing threats sent to thousands of schools in Italy, Scotland, Ireland and the UK and to about 20 organisations including newsrooms and airports, from compromised accounts at Taiwan's Da-Yeh University, a Bogotá health agency and the Philippines' Department of Education. Irish police think they came from abroad to "sow discord" around the European Political Community summit in Dublin; Proofpoint says some aspects could fit a nation-state-aligned actor, but nothing is attributed. The Record (source)

Outages & Resilience 1

🟡 Worth knowing

Microsoft 365 recovers after a day-long Outlook and Teams outage

Added after publishingMicrosoft 365OutlookTeamsoutage

WhatA day-long Microsoft 365 outage hit Outlook and Teams worldwide, blamed on a network capacity upgrade issue, not malicious activity.

Details and sources
Microsoft first reported the problem shortly before 1 p.m. ET on 12 November; Outlook, Teams and other Microsoft 365 services were disrupted for businesses, governments, schools and healthcare providers worldwide (100,000+ Downdetector reports within an hour). Microsoft blames "an issue during an earlier network capacity upgrade" and says there is "no indication this is the result of malicious or cyber activity"; by the morning of 13 November it reported recovery in North America, Europe and Asia Pacific, with mail queues nearing normal and some delayed delivery. ABC News (source) · The Spokesman-Review (source)

Privacy & Regulation 1

⚪ Context

New York City Council votes to ban surveillance pricing

Added after publishingNew York Citysurveillance pricingban

WhatThe New York City Council voted to ban using personal data and algorithms to charge different people different prices; it awaits the mayor's signature.

Details and sources
on 12 November the Council passed Introduction 891-A, which bans using personal data and algorithms to give different people different prices for the same goods or services, and restricts collecting, selling or sharing personal data for that purpose; loyalty programmes and discounts for veterans, seniors and teachers stay allowed under conditions. It awaits the mayor's signature and would take effect a year after becoming law; the article gives no vote count or penalties. Malwarebytes (source)
‹ 12/11/202616/11/2026 ›