<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<title>CyberView: high-priority cyber events</title>
<link>https://xanderhutch.github.io/CyberView/</link>
<atom:link href="https://xanderhutch.github.io/CyberView/data/feed.xml" rel="self" type="application/rss+xml"/>
<description>High-severity incidents, exploited vulnerabilities and urgent advisories from CyberView, with what to do, researched and written by CyberView AI (a Claude model). Unverified claims are left out.</description>
<language>en</language>
<lastBuildDate>Sun, 22 Nov 2026 15:32:34 GMT</lastBuildDate>
<item>
<title>KEPCO says personal data of over one and a half million people leaked from its customer and procurement systems</title>
<link>https://xanderhutch.github.io/CyberView/#e=kepco-hanjeon-on-srm-personal-data-leak-2026-11</link>
<guid isPermaLink="false">cyberview:kepco-hanjeon-on-srm-personal-data-leak-2026-11</guid>
<pubDate>Sun, 22 Nov 2026 10:07:58 GMT</pubDate>
<category>incident</category>
<description>KEPCO said on 22 November it confirmed on 20 November a personal data leak from its Hanjeon ON customer platform and SRM procurement system: 152만248 people (over one and a half million) by simple sum, Newsis reports, overlap unchecked. Exposed: resident registration, bank account and card numbers. Dailian reports KEPCO presumes an outside attacker and has told police; it blocked external IP access and is investigating with the National Intelligence Service.

What to do: KEPCO urges caution with calls, texts and emails that impersonate KEPCO or come from unclear sources, and runs a dedicated service where customers can check whether their data leaked and report damage. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Poland&apos;s ePUAP e-government platform restricted after cyberattack; full operation restored 19 November, COI says</title>
<link>https://xanderhutch.github.io/CyberView/#e=epuap-cyberattack-poland-2026-11</link>
<guid isPermaLink="false">cyberview:epuap-cyberattack-poland-2026-11</guid>
<pubDate>Sat, 21 Nov 2026 08:30:25 GMT</pubDate>
<category>incident</category>
<description>Poland’s COI told Sekurak it preventively restricted access to part of the ePUAP system from 10 November, that the attack was detected and stopped, and that full operation returned on 19 November. Deputy PM Krzysztof Gawkowski said the attack came from abroad and that logs showed preparatory activity for a December operation against public administration. COI says the ABW is investigating and it found no sign of spread.

What to do: No action needed; for awareness. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>ZOZO: about 15.18 million ZOZOTOWN customers&apos; data leaked via PLAID KARTE breach</title>
<link>https://xanderhutch.github.io/CyberView/#e=zozo-karte-customer-data-leak-2026-11</link>
<guid isPermaLink="false">cyberview:zozo-karte-customer-data-leak-2026-11</guid>
<pubDate>Fri, 20 Nov 2026 23:30:00 GMT</pubDate>
<category>incident</category>
<description>ZOZO said on 20 November that customer information it managed on PLAID&apos;s KARTE leaked. About 15.18 million ZOZOTOWN customers incl. former members (member ID, age, gender, postal code) are affected, about 47,000 of them also with login name, email and phone number; about 13.93 million WEAR by ZOZO customers; about 24,000 brand staff. Passwords are not managed in KARTE; card data is not confirmed leaked.

What to do: ZOZO customers: be wary of phishing emails, texts or calls impersonating ZOZO, and do not open their links or attachments, as ZOZO advises. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>SMBC says up to about 14 million customer records leaked via PLAID KARTE breach</title>
<link>https://xanderhutch.github.io/CyberView/#e=smbc-karte-customer-data-leak-2026-11</link>
<guid isPermaLink="false">cyberview:smbc-karte-customer-data-leak-2026-11</guid>
<pubDate>Fri, 20 Nov 2026 23:30:00 GMT</pubDate>
<category>incident</category>
<description>SMBC said on 20 November that up to about 14 million records of customer information leaked after unauthorized access to KARTE, a service from its contractor PLAID. The data is dates of birth, category values for age band, gender and whether a phone number is registered, and notice flags; SMBC says it is not linked to names and cannot identify individuals. No impact on assets or transactions is confirmed so far.

What to do: SMBC says it will never ask for passwords or card details by email or phone about this; it urges customers not to open links or attachments in suspicious messages claiming to be from the bank. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Marui Group: about 8.83 million and about 8.5 million customer records leaked via PLAID KARTE breach</title>
<link>https://xanderhutch.github.io/CyberView/#e=marui-group-karte-customer-data-leak-2026-11</link>
<guid isPermaLink="false">cyberview:marui-group-karte-customer-data-leak-2026-11</guid>
<pubDate>Fri, 20 Nov 2026 23:30:00 GMT</pubDate>
<category>incident</category>
<description>Marui Group said on 20 November that customer data leaked after unauthorized access to PLAID&apos;s KARTE, which it says came from PLAID&apos;s system, not Marui&apos;s. About 8.83 million people (name, gender, date of birth, email, customer management number) and about 8.5 million EPOS CARD users incl. former members (customer management number, gender) are affected; about 60,000 of them may also have name, age, annual income and cash-advance limit exposed. Card numbers are not confirmed leaked.

What to do: Marui and EPOS CARD customers: be careful with suspicious emails and do not open URLs in them or enter personal information, as Marui Group advises. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>React2STEAL: GitHub Actions scanners exploit React2Shell and publish stolen secrets</title>
<link>https://xanderhutch.github.io/CyberView/#e=react2steal-github-actions-campaign-2026-11</link>
<guid isPermaLink="false">cyberview:react2steal-github-actions-campaign-2026-11</guid>
<pubDate>Fri, 20 Nov 2026 10:05:58 GMT</pubDate>
<category>incident</category>
<description>OpenSourceMalware published research on November 18 on a campaign it named React2STEAL: 45 GitHub repositories across 44 accounts run scanners for React2Shell (CVE-2025-55182) and CVE-2025-29927 and commit results to public repositories. It found at least 1,281 victim hostnames across about 1,100 domains. It lists the actor as unknown. GitGuardian&apos;s write-up says patches for both flaws have been available for months.

What to do: Upgrade React and Next.js to fixed versions; until then strip or reject the x-middleware-subrequest header at the proxy or WAF, check logs for it and for /_next/image requests aimed at the metadata address, rotate every secret the server could reach, and enforce IMDSv2.

CVEs: CVE-2025-55182, CVE-2025-29927 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Huntress: attackers exploit ScreenConnect flaw CVE-2026-84869 against MSP technicians&apos; workstations</title>
<link>https://xanderhutch.github.io/CyberView/#e=huntress-screenconnect-cve-2026-84869-technician-exploitation-2026-11</link>
<guid isPermaLink="false">cyberview:huntress-screenconnect-cve-2026-84869-technician-exploitation-2026-11</guid>
<pubDate>Fri, 20 Nov 2026 08:30:12 GMT</pubDate>
<category>vuln</category>
<description>Huntress reports attackers exploiting CVE-2026-84869 in ScreenConnect, which lets a guest push files that run on a technician&apos;s machine. It saw one full compromise of an MSP technician workstation on November 12 and, since October 20, at least 688,000 guest &apos;run file&apos; requests for update*.bat hitting 42 MSP servers from 29 IP addresses. Huntress recommends updating technician host clients to 26.6.5 or later; updating the server alone does not fix it.

What to do: Update every technician&apos;s ScreenConnect host client to 26.6.5 or later; if you cannot, remove the TransferFiles permission from roles and session groups, as ConnectWise advises, and check Session.db for guest RanFiles or SentFiles events.

CVEs: CVE-2026-84869 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Santos port authority systems still partly down after 6 November cyberattack; Federal Police investigating</title>
<link>https://xanderhutch.github.io/CyberView/#e=brazil-santos-port-authority-ransom-demand-2026-11</link>
<guid isPermaLink="false">cyberview:brazil-santos-port-authority-ransom-demand-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 17:40:00 GMT</pubDate>
<category>incident</category>
<description>Santos Port Authority (APS) systems went down on 6 November. Gazeta do Povo (13 November) reports a Federal Police inquiry opened on 10 November and that hackers emailed APS seeking contact, unanswered. ABC do ABC (18 November) says more than 30 systems still run in alternative mode; APS cites a power failure then an intrusion that erased data, and says no confidential data leaked and cargo handling was never compromised.

What to do: General advice: port users and shippers should rely only on channels announced by the Autoridade Portuária de Santos, and treat unexpected emails or calls about cargo, access or payments with caution. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Google: ARMORY CASTLE exploits vCenter zero-day CVE-2026-123890 in espionage campaign</title>
<link>https://xanderhutch.github.io/CyberView/#e=gtig-armory-castle-vcenter-zero-day-2026-11</link>
<guid isPermaLink="false">cyberview:gtig-armory-castle-vcenter-zero-day-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 14:50:00 GMT</pubDate>
<category>vuln</category>
<description>Google Threat Intelligence Group says ARMORY CASTLE, a suspected PRC-nexus espionage actor it formerly tracked as UNC5221, exploited CVE-2026-123890, an unauthenticated authentication bypass in VMware vCenter Server, from early September 2026. SecurityWeek reports a CVSS of 9.8 and fixes in vCenter 8.0 U3m, 9.0.2.0200 and 9.1.1.0200, with no public patch for 7.0. Google names no victims.

What to do: Google advises applying Broadcom&apos;s patches immediately; SecurityWeek lists fixed vCenter builds 8.0 U3m, 9.0.2.0200 and 9.1.1.0200 and says 7.0 has no public patch. Google also advises restricting inbound access to and egress from vCenter, as described in its post.

CVEs: CVE-2026-123890, CVE-2026-102764 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Cisco Expressway clustering flaw CVE-2026-76466 exploited after public PoC, F6 reports</title>
<link>https://xanderhutch.github.io/CyberView/#e=cisco-expressway-cve-2026-76466-exploited-2026-11</link>
<guid isPermaLink="false">cyberview:cisco-expressway-cve-2026-76466-exploited-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 13:40:00 GMT</pubDate>
<category>vuln</category>
<description>Cisco&apos;s critical advisory (CVSS 9.8) says an unauthenticated remote attacker can run commands as root on clustered Expressway-C and Expressway-E. SecurityWeek reports that the firm F6 investigated an attack on two internet-exposed X15.5.3 devices that spread to two cluster nodes, after a PoC was posted on GitHub; Security Affairs says the victim was a Russian organization. Cisco&apos;s advisory, unrevised, still says it knows of no malicious use. Fix: X15.5.20.

What to do: If you run clustered Expressway, upgrade to X15.5.20 now (X14 and earlier must move to a fixed release; Cisco says no workaround). If you cannot patch at once, F6 advises blocking TCP 4371-4372 except from trusted cluster node IPs.

CVEs: CVE-2026-76466 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>SonicWall: retire Gen 6 firewalls now, no more patches even for critical flaws</title>
<link>https://xanderhutch.github.io/CyberView/#e=sonicwall-gen6-end-of-support-retire-2026-11</link>
<guid isPermaLink="false">cyberview:sonicwall-gen6-end-of-support-retire-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 10:53:30 GMT</pubDate>
<category>vuln</category>
<description>SonicWall&apos;s advisory of 18 November says every Gen 6 and Gen 6.5 firewall running SonicOS 6.x, including NSv, has reached end of support and will get no firmware or security patches regardless of severity. SonicWall says it expects internet-facing Gen 6 units to be actively targeted and tells customers to replace them with Gen 8. No new CVE is named.

What to do: Inventory every Gen 6 unit; replace internet-facing ones first; meanwhile disable WAN management and SSL VPN (or restrict sources), enforce MFA, and monitor logs. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>WordPress 7.1.5 security release fixes unauthenticated privilege escalation and SQL injection</title>
<link>https://xanderhutch.github.io/CyberView/#e=wordpress-7-1-5-security-release-2026-11</link>
<guid isPermaLink="false">cyberview:wordpress-7-1-5-security-release-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 05:54:00 GMT</pubDate>
<category>vuln</category>
<description>WordPress 7.1.5, released November 18, is a security release listing 20 items, including unauthenticated privilege escalation through Template Part block attributes that the project says could lead to remote code execution. It also fixes two SQL injections, a Multisite Super Admin takeover and stored XSS in comments. The project recommends updating immediately and does not report exploitation.

What to do: Update WordPress to 7.1.5 now (Dashboard, Updates), or confirm the background auto-update ran; older branches are being backported, so watch for those releases. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Bulgaria says hackers had access to government networks for about five years</title>
<link>https://xanderhutch.github.io/CyberView/#e=bulgaria-government-networks-five-year-espionage-2026-11</link>
<guid isPermaLink="false">cyberview:bulgaria-government-networks-five-year-espionage-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 04:40:00 GMT</pubDate>
<category>incident</category>
<description>Bulgaria&apos;s digital minister Ivan Vassilev said that since 2021 more than 130 GB, or 62 million pages, of documents had been downloaded from state networks. Reuters reports he said several international hacking groups were believed responsible, with no conclusive evidence linking them to a foreign government. A new monitoring system flagged the activity; access was cut and malware removed. No evidence so far that personal data was compromised.

What to do: No action needed; for awareness. Government and critical-sector defenders should review document-exchange systems for long-dwell access and centralise monitoring. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>DHC: about 19.11 million customer records leaked via PLAID KARTE breach</title>
<link>https://xanderhutch.github.io/CyberView/#e=dhc-karte-19-million-customer-data-leak-2026-11</link>
<guid isPermaLink="false">cyberview:dhc-karte-19-million-customer-data-leak-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 04:25:00 GMT</pubDate>
<category>incident</category>
<description>DHC announced on November 18 that about 19.11 million online-shop customer records leaked after unauthorized access to PLAID&apos;s KARTE marketing service. NHK, Kyodo and Yomiuri report names, email addresses and dates of birth; Kyodo and Yomiuri say credit card payment data was not included, and Yomiuri says no misuse has been confirmed.

What to do: DHC customers: treat unexpected emails or texts using your name and birth date as phishing; change reused passwords. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Rakuten Securities says test-environment data on up to about 4.47 million accounts may have leaked</title>
<link>https://xanderhutch.github.io/CyberView/#e=rakuten-securities-test-environment-breach-2026-11</link>
<guid isPermaLink="false">cyberview:rakuten-securities-test-environment-breach-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 04:08:00 GMT</pubDate>
<category>incident</category>
<description>Rakuten Securities said it learned on November 16 that a third party may have accessed its systems. Test-environment data may have leaked on November 11: details of up to about 4.47 million accounts, and customer codes and login IDs for up to about 11.25 million. No password or PIN leak or unauthorised trades were found as of November 17. On November 19 the company said it received a report-collection order from Japan&apos;s Financial Services Agency.

What to do: Rakuten Securities customers: be wary of messages or calls posing as the company or its partners, never share passwords or PINs, and watch for the company&apos;s notice. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Chile&apos;s Huatacondo solar plant still offline almost three months after cyberattack</title>
<link>https://xanderhutch.github.io/CyberView/#e=huatacondo-solar-bess-cyberattack-offline-2026-11</link>
<guid isPermaLink="false">cyberview:huatacondo-solar-bess-cyberattack-offline-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 04:02:27 GMT</pubDate>
<category>incident</category>
<description>Chile&apos;s Coordinador Eléctrico Nacional recorded the forced disconnection of the 103 MW Huatacondo solar plant and its 98 MW/312 MWh battery system on 27 August with the cause &apos;cyberattack&apos;, and its 14 November report still showed both at 0% availability. The Coordinador says it activated disconnection protocols after the company reported the event. On 11 November a deputy asked the security and energy ministries for details. No technical cause or reconnection date has been published.

What to do: Operators of solar, storage and thermal plants: confirm remote access paths and telecontrol are monitored and that incident-reporting steps under Chile&apos;s Law 21.663 are rehearsed. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Drupal reports core flaw CVE-2026-132165 exploited on live sites; fixed in 10.6.22, 11.3.22 and 11.4.13</title>
<link>https://xanderhutch.github.io/CyberView/#e=drupal-core-installer-rce-sa-core-2026-017-2026-11</link>
<guid isPermaLink="false">cyberview:drupal-core-installer-rce-sa-core-2026-017-2026-11</guid>
<pubDate>Thu, 19 Nov 2026 00:14:29 GMT</pubDate>
<category>vuln</category>
<description>Drupal rates SA-CORE-2026-017, CVE-2026-132165, highly critical: under specific conditions install.php can be used to reinstall a site, leading to takeover or code execution. In an update to the advisory, Drupal says its Security Team received a report this was exploited on live sites as early as 2026-11-19 19:00 UTC. Fixed in 10.6.22, 11.3.22 and 11.4.13. A second advisory, CVE-2026-132166, is a gadget chain that is not directly exploitable.

What to do: Drupal advises updating core to 10.6.22, 11.3.22 or 11.4.13; if you cannot upgrade, secure the installer as Drupal describes. Older branches are end of life, per Drupal.

CVEs: CVE-2026-132165, CVE-2026-132166 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CISA adds four exploited flaws to KEV: Synology DSM, LiteLLM, Apache Axis, REDCap</title>
<link>https://xanderhutch.github.io/CyberView/#e=cisa-kev-four-additions-2026-11-18</link>
<guid isPermaLink="false">cyberview:cisa-kev-four-additions-2026-11-18</guid>
<pubDate>Thu, 19 Nov 2026 00:14:29 GMT</pubDate>
<category>vuln</category>
<description>CISA added four vulnerabilities to its Known Exploited Vulnerabilities catalog on November 18, citing evidence of active exploitation: CVE-2024-10441 in Synology DiskStation Manager, CVE-2026-40217 in BerriAI LiteLLM (unprotected alternate channel), CVE-2026-90436 in Apache Axis (unsafe reflection) and CVE-2026-90817 in Vanderbilt University REDCap (code injection). The alert gives no fixed versions. The Hacker News reports exploitation including ransomware on Synology NAS, cryptomining and LLMjacking.

What to do: CISA directs federal agencies to remediate under BOD 26-04: find Synology DSM, LiteLLM, Apache Axis and REDCap in your environment and apply vendor fixes where they exist. The Hacker News reports Axis 1.x is end of life and unpatched, and names migrating or applying the July 12, 2026 commit as options.

CVEs: CVE-2024-10441, CVE-2026-40217, CVE-2026-90436, CVE-2026-90817 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>SANS honeypots see scans for chained LiteLLM and Starlette flaws (CVE-2026-42271, CVE-2026-48710)</title>
<link>https://xanderhutch.github.io/CyberView/#e=sans-isc-litellm-starlette-scanning-2026-11</link>
<guid isPermaLink="false">cyberview:sans-isc-litellm-starlette-scanning-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 22:11:48 GMT</pubDate>
<category>vuln</category>
<description>SANS ISC honeypots recorded requests to /mcp-rest/test/tools/list attempting to chain CVE-2026-42271 in LiteLLM with CVE-2026-48710 in Starlette for unauthenticated remote code execution. Starlette before 1.0.1 does not validate the Host header. SANS cites a Horizon3.ai write-up and says exploitation has been reported since July. Whether any attempts succeeded is not stated.

What to do: Upgrade Starlette to 1.0.1 or later, update LiteLLM, keep LiteLLM off the public internet, and rotate provider keys on any exposed instance.

CVEs: CVE-2026-42271, CVE-2026-48710 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CERT-FR: NGINX flaws allow unauthenticated remote code execution over HTTP/2; public PoC exists</title>
<link>https://xanderhutch.github.io/CyberView/#e=cert-fr-nginx-http2-rce-2026-11</link>
<guid isPermaLink="false">cyberview:cert-fr-nginx-http2-rce-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 21:13:25 GMT</pubDate>
<category>vuln</category>
<description>CERT-FR warns that two flaws in F5 NGINX, CVE-2026-123633 and CVE-2026-123007, allow unauthenticated remote code execution over HTTP/2; a public proof of concept exists for CVE-2026-123007. Fixed in Open Source 1.31.9 and 1.30.7. Palo Alto Networks says PAN-OS and Panorama with HTTP/2 enabled are affected and it is not aware of malicious exploitation. Disabling HTTP/2 is an interim workaround for those two flaws.

What to do: Check every NGINX Open Source and NGINX Plus server that serves HTTP/2. Apply the fixed versions from F5 bulletins K000163926 and K000163928, or disable HTTP/2 until patched (this covers only the HTTP/2 flaws; the resolver, HTTP/3 and mail proxy flaws need the fixed versions, 1.30.7 or 1.31.9) (remove http2 from the listen line, or set http2 off on 1.25.0 and earlier). Review exposed servers for signs of compromise.

CVEs: CVE-2026-123633, CVE-2026-123007, CVE-2026-123634, CVE-2026-121810, CVE-2026-113425 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Microsoft Agent Framework PyPI package compromised with credential stealer</title>
<link>https://xanderhutch.github.io/CyberView/#e=microsoft-agent-framework-pypi-compromise-2026-11</link>
<guid isPermaLink="false">cyberview:microsoft-agent-framework-pypi-compromise-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 18:14:23 GMT</pubDate>
<category>incident</category>
<description>Versions 1.26.1 and 1.26.2 of the agent-framework package on PyPI were published on November 17 with a credential stealer that runs when Python starts, targeting cloud, GitHub and AI credentials and tokens. PyPI quarantined both; 1.26.0 is the last safe version. Italy&apos;s CSIRT (ACN) says the malware installed no OS-level persistence and downloaded no further payloads. As of 18 November, Microsoft had not responded to BleepingComputer.

What to do: Check for Agent Framework 1.26.1 or 1.26.2 in environments and CI builds, pin to 1.26.0, and rotate Azure, AWS, GitHub, Docker, Hugging Face and OpenAI credentials and OIDC tokens on any host that installed them. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Apache warns Roller RCE flaw (CVE-2026-129448) is exploited in attacks</title>
<link>https://xanderhutch.github.io/CyberView/#e=apache-roller-rce-cve-2026-129448-2026-11</link>
<guid isPermaLink="false">cyberview:apache-roller-rce-cve-2026-129448-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 18:14:23 GMT</pubDate>
<category>vuln</category>
<description>Apache says an unauthenticated remote code execution flaw in Roller, CVE-2026-129448, is being exploited. It is a deserialization bug in the XML-RPC interface affecting versions 4.0.0 through 6.1.4, and is the same bug as CVE-2026-82384, which named only 6.1.5. Version 6.1.6 or later fixes it.

What to do: Upgrade Apache Roller to 6.1.6 or later. If you cannot, Apache&apos;s workaround is to set enabledForExtensions to false in WEB-INF/web.xml or remove the /roller-services/xmlrpc mapping (reports differ on whether both are needed), then restart; switching off XML-RPC in the admin UI is not enough.

CVEs: CVE-2026-129448 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>NCSC Netherlands: public exploit code for critical Ivanti Endpoint Manager flaw (CVE-2026-121653)</title>
<link>https://xanderhutch.github.io/CyberView/#e=ncsc-nl-ivanti-endpoint-manager-sql-injection-2026-11</link>
<guid isPermaLink="false">cyberview:ncsc-nl-ivanti-endpoint-manager-sql-injection-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:50:00 GMT</pubDate>
<category>vuln</category>
<description>The Dutch National Cyber Security Centre warns that proof-of-concept code is now public for CVE-2026-121653, an SQL injection flaw in the Core of Ivanti Endpoint Manager before 2024 SU8. A remote attacker without credentials can use it to run code. A second flaw, CVE-2026-18128, lets an attacker positioned on the network read and alter agent traffic sent in clear text. Ivanti has released updates.

What to do: Install the updates Ivanti has released for Endpoint Manager, as the NCSC-NL advisory advises; see Ivanti&apos;s November 2026 security advisory for affected versions and fixes.

CVEs: CVE-2026-121653, CVE-2026-18128 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Japan&apos;s RIKEN takes Fugaku supercomputer offline after intrusion</title>
<link>https://xanderhutch.github.io/CyberView/#e=riken-fugaku-supercomputer-cyberattack-2026-11</link>
<guid isPermaLink="false">cyberview:riken-fugaku-supercomputer-cyberattack-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:25:00 GMT</pubDate>
<category>incident</category>
<description>RIKEN&apos;s first notice says an outside party reached Fugaku&apos;s internal operational network from November 10 to 13 through a public web page showing Fugaku&apos;s operational status; the exact method is under investigation. Its second notice (20 Nov) says 4,222 Fugaku users&apos; data was taken and 430 credentials (233 SSH private keys without a passphrase, 197 with one) were exfiltrated, some used for external connections. Fugaku stays offline; no attacker is named.

What to do: Fugaku users: revoke and replace SSH keys used with Fugaku, as RIKEN asks, and watch for phishing emails. Others: no action needed; for awareness. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Canadian Centre for Cyber Security: Microsoft November rollup AV26-1142 updated for exploited flaw CVE-2026-124194</title>
<link>https://xanderhutch.github.io/CyberView/#e=cccs-microsoft-november-rollup-advisory-av26-1142-2026-11</link>
<guid isPermaLink="false">cyberview:cccs-microsoft-november-rollup-advisory-av26-1142-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>Cyber Centre advisory AV26-1142 relays Microsoft&apos;s November 2026 security updates to Canadian organisations. The affected list spans Windows 10 and 11, Windows Server 2012 to 2025, Microsoft 365 Apps and Office, Exchange Server Subscription Edition, SharePoint Server, SQL Server, .NET, Visual Studio and many Azure services. Update 1, dated November 12, adds that Microsoft reports CVE-2026-124194 as exploited and that CISA catalogued it on November 10.

What to do: Install Microsoft&apos;s November 2026 security updates across Windows, Office, Exchange, SharePoint, SQL Server and developer tools, taking the fix for CVE-2026-124194 first. Use the MSRC release notes linked from the advisory to scope the work.

CVEs: CVE-2026-124194 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>NCSC New Zealand alerts on OpenSSH &apos;SSHole&apos; authentication bypass CVE-2026-123158</title>
<link>https://xanderhutch.github.io/CyberView/#e=ncsc-nz-openssh-sshole-cve-2026-123158-2026-11</link>
<guid isPermaLink="false">cyberview:ncsc-nz-openssh-sshole-cve-2026-123158-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>New Zealand&apos;s NCSC issued an alert on November 11 about CVE-2026-123158, nicknamed SSHole, an authentication bypass in OpenSSH on Linux that may give an attacker root access. A system is exposed only if it runs a version before 10.6p2, uses musl libc, does not use PAM and accepts password logins. The alert does not say whether the flaw is being exploited.

What to do: Identify Linux systems running OpenSSH before 10.6p2 on musl libc without PAM that accept password logins, and update to 10.6p2. Until then, disable password authentication or turn SSH off on those systems.

CVEs: CVE-2026-123158 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CERT-EU rates Citrix NetScaler flaws critical: unauthenticated memory overflows scored up to 9.5</title>
<link>https://xanderhutch.github.io/CyberView/#e=cert-eu-citrix-netscaler-advisory-2026-020-2026-11</link>
<guid isPermaLink="false">cyberview:cert-eu-citrix-netscaler-advisory-2026-020-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>CERT-EU Security Advisory 2026-020 covers 28 memory flaws in Citrix NetScaler ADC and NetScaler Gateway. CVE-2026-120868, scored 9.3, sits in core packet processing and affects every deployment, default configuration included, without authentication. CVE-2026-107412, scored 9.5, affects appliances set up as a Gateway or AAA virtual server. Citrix reports no exploitation and no public proof of concept is confirmed.

What to do: Upgrade customer-managed NetScaler ADC and Gateway to 14.1-73.57 or 13.1-64.39, FIPS builds to 14.1-73.58 FIPS or 13.1-37.293. Do internet-facing Gateway and AAA appliances first.

CVEs: CVE-2026-120868, CVE-2026-107412 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Canadian Centre for Cyber Security: Check Point advisory AV26-1154 updated to note an exploited flaw</title>
<link>https://xanderhutch.github.io/CyberView/#e=cccs-check-point-advisory-av26-1154-2026-11</link>
<guid isPermaLink="false">cyberview:cccs-check-point-advisory-av26-1154-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>Cyber Centre advisory AV26-1154, issued November 12 and updated the next day, covers two flaws across multiple versions of Check Point Force firewalls, management and log servers, Multi-Domain products and Maestro Orchestrator. It relays Check Point&apos;s statement that CVE-2026-123039, a local escalation to root, is being exploited in the wild. Update 1 records that CISA added that flaw to its exploited-vulnerabilities catalog on November 13.

What to do: Apply Check Point&apos;s fixes for CVE-2026-123039 and CVE-2026-107484 on firewalls, management and log servers, Multi-Domain servers and Maestro Orchestrator. Read the Check Point advisories and the CISA catalog entry linked from the advisory.

CVEs: CVE-2026-123039, CVE-2026-107484 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CSA Singapore warns of active exploitation of Apache HTTP Server mod_http2 flaw CVE-2026-119618</title>
<link>https://xanderhutch.github.io/CyberView/#e=csa-singapore-apache-mod-http2-exploitation-al-2026-164-2026-11</link>
<guid isPermaLink="false">cyberview:csa-singapore-apache-mod-http2-exploitation-al-2026-164-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>Singapore&apos;s Cyber Security Agency issued alert AL-2026-164 on November 17 about CVE-2026-119618, a flaw rated 8.1 in the mod_http2 module of Apache HTTP Server. Versions 2.4.55 up to but not including 2.4.71 are affected when that module is switched on. The agency says exploitation is reportedly under way and that proof-of-concept code is public.

What to do: Find Apache HTTP Server 2.4.55 to 2.4.70 with mod_http2 enabled and update to the latest release at once, internet-facing servers first. Check the Apache advisory for the full list of affected versions.

CVEs: CVE-2026-119618 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CISA adds exploited Perforce Puppet Enterprise flaw (CVE-2026-115731) to KEV</title>
<link>https://xanderhutch.github.io/CyberView/#e=cisa-kev-puppet-enterprise-cve-2026-115731-2026-11</link>
<guid isPermaLink="false">cyberview:cisa-kev-puppet-enterprise-cve-2026-115731-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:20:00 GMT</pubDate>
<category>vuln</category>
<description>CISA added CVE-2026-115731, a deserialization flaw in Perforce Puppet Enterprise, to its KEV catalog on November 17, citing active exploitation; federal agencies must fix it by December 1. Help Net Security reports a client holding a Puppet CA-signed certificate can send a crafted report and run code as the Puppet Server user. Fixed in 2026.0.1, 2025.12.1 and 2023.8.13.

What to do: Upgrade Puppet Enterprise to 2026.0.1, 2025.12.1 or 2023.8.13 (Puppet Core 9.1.3 or 8.22.2) and restart the server. Review autosign settings and look for unrecognised signed node certificates.

CVEs: CVE-2026-115731 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Dragos counts 1,266 ransomware incidents at industrial organisations in Q3 2026</title>
<link>https://xanderhutch.github.io/CyberView/#e=dragos-industrial-ransomware-q3-2026-11</link>
<guid isPermaLink="false">cyberview:dragos-industrial-ransomware-q3-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:08:00 GMT</pubDate>
<category>report</category>
<description>Dragos recorded 1,266 ransomware incidents affecting industrial organisations in the third quarter of 2026, up 11% from 1,140 the quarter before. Manufacturing took 69% of them and the United States 37%. According to Dragos, Qilin led with 195 incidents, ahead of The Gentlemen (167 claimed victims) and Cl0p (35). Dragos names exploited edge devices and enterprise applications, and stolen VPN credentials, among the ways in. Dragos saw no operator directly manipulate an industrial control system.

What to do: Rotate administrator and VPN credentials, especially on FortiGate appliances, and enforce multi-factor sign-in on all remote access. Audit for administrator accounts you did not create and patch edge devices quickly.

CVEs: CVE-2026-0257, CVE-2026-15409, CVE-2026-15410, CVE-2026-63077, CVE-2026-59310, CVE-2026-12569 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Rapid7 and CISA: critical Java flaw CVE-2026-114938 (Java4Shell) in Java 17 and later; Oracle patches released</title>
<link>https://xanderhutch.github.io/CyberView/#e=rapid7-java-17-rce-cve-2026-114938-2026-11</link>
<guid isPermaLink="false">cyberview:rapid7-java-17-rce-cve-2026-114938-2026-11</guid>
<pubDate>Wed, 18 Nov 2026 00:08:00 GMT</pubDate>
<category>vuln</category>
<description>Rapid7 and CISA warn of CVE-2026-114938 (Java4Shell), a memory corruption bug in the x86-64 HotSpot JVM Base64 decoder that could allow remote code execution; Oracle released fixes on November 17. SecurityWeek (Nov 19) reports GreyNoise saw over 150 matching requests from five IP addresses, with no successful intrusions seen on its sensors; Cisco, Siemens, Elastic and Red Hat issued advisories.

What to do: Inventory every Java runtime, including bundled ones, and install the fixed releases (for example 17.0.21.1, 21.0.13.1, 25.0.5.1, 27.0.1.1), internet-facing systems first. See Rapid7&apos;s advisory for compensating controls where patching is not possible.

CVEs: CVE-2026-114938 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>MS-ISAC: SonicWall SonicOS flaws allow authentication bypass; one SSLVPN bug is already on CISA&apos;s KEV list</title>
<link>https://xanderhutch.github.io/CyberView/#e=ms-isac-sonicwall-sonicos-auth-bypass-2026-11</link>
<guid isPermaLink="false">cyberview:ms-isac-sonicwall-sonicos-auth-bypass-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:38:00 GMT</pubDate>
<category>vuln</category>
<description>MS-ISAC advisory 2026-126 covers seven flaws in SonicWall SonicOS on Gen7 firewalls up to 7.3.3-7015 and Gen8 up to 8.2.2-8015. Two management-interface bugs let remote unauthenticated attackers perform unauthorised operations under certain conditions. CVE-2026-19868 lets a user with other valid credentials skip SAML, and possibly MFA, on SSLVPN; SonicWall investigated a report of exploitation and CISA listed it as exploited on November 6.

What to do: Apply SonicWall&apos;s SonicOS updates for Gen7 and Gen8 firewalls after testing, starting with devices using SAML for SSLVPN, and keep firewall management interfaces off the internet.

CVEs: CVE-2026-107154, CVE-2026-107155, CVE-2026-19703, CVE-2026-19868, CVE-2026-82308, CVE-2026-77708, CVE-2026-77709 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>FBI: FLASH details North Korean FaceStealer fake-meeting scheme behind $400M in crypto theft</title>
<link>https://xanderhutch.github.io/CyberView/#e=fbi-flash-facestealer-fake-meetings-2026-11</link>
<guid isPermaLink="false">cyberview:fbi-flash-facestealer-fake-meetings-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:38:00 GMT</pubDate>
<category>report</category>
<description>FBI FLASH AC-111226-MS (TLP:CLEAR) attributes the FaceStealer scheme to North Korean state-sponsored hackers (Sapphire Sleet, UNC1069, BlueNoroff). They hijack a victim&apos;s messaging account, invite contacts to a counterfeit Zoom or Teams call, replay recorded webcam video and push a fake audio fix that installs malware. The FBI says over $400 million was stolen in about 18 months.

What to do: Join calls through the official Zoom or Teams app with an invite code, keep the webcam off until the host is verified another way, never run scripts or paste code to fix audio, and enable MFA and login alerts on messaging accounts. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CISA: six industrial control advisories cover Siemens building, edge and RUGGEDCOM products</title>
<link>https://xanderhutch.github.io/CyberView/#e=cisa-ics-advisories-siemens-ruggedcom-2026-11</link>
<guid isPermaLink="false">cyberview:cisa-ics-advisories-siemens-ruggedcom-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:38:00 GMT</pubDate>
<category>vuln</category>
<description>CISA released six industrial control system advisories on November 12: five for Siemens products (building automation and energy, Desigo and Gridscale, Industrial Edge Management and two for RUGGEDCOM APE1808) and an update for Hitachi Energy RTU500. One RUGGEDCOM advisory covers a Brotli decoder flaw in the bundled PAN-OS, rated 9.8, that a remote unauthenticated attacker could use to crash the device or possibly run code.

What to do: Operators of the listed Siemens and Hitachi Energy products: read the matching advisory. For RUGGEDCOM APE1808, upgrade PAN-OS to a corrected version, restrict network access to the device and keep it off the internet.

CVEs: CVE-2026-111497 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Caixa Vida e Previdência breach exposes data of 7.6 million clients in Brazil</title>
<link>https://xanderhutch.github.io/CyberView/#e=caixa-vida-previdencia-data-breach-2026-11</link>
<guid isPermaLink="false">cyberview:caixa-vida-previdencia-data-breach-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:02:00 GMT</pubDate>
<category>incident</category>
<description>The insurer and pension company, part of Caixa Seguridade, said on November 12 it detected unauthorized access on November 6 and that registration data was copied. Per Estadão Conteúdo, the base covers all 7.6 million active clients plus ex-clients; names and CPFs for all records, other details for part. It says no financial or authentication data appears affected, and it notified ANPD on November 11.

What to do: If you are a Caixa Vida e Previdência client, be wary of calls, emails or messages that cite your CPF or policy. Do not share codes or passwords, and use the company&apos;s official channel for questions. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>eMudhra revokes eight more CAs as Microsoft pulls trust in all its roots</title>
<link>https://xanderhutch.github.io/CyberView/#e=emudhra-ca-revocations-microsoft-distrust-2026-11</link>
<guid isPermaLink="false">cyberview:emudhra-ca-revocations-microsoft-distrust-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:02:00 GMT</pubDate>
<category>incident</category>
<description>The Indian certificate authority revoked eight more intermediate CAs after tracing weak keys from its October compromise to one line in a home-made script that cut ECDSA private values to a single byte. Microsoft removes all 10 remaining eMudhra roots from Windows, Edge and Azure from November 16; Apple ends trust on November 30. eMudhra is leaving public TLS and has withdrawn its assurance that RSA issuers were unaffected.

What to do: Inventory certificates chaining to eMudhra roots and replace them with certificates from another CA now. Apply current Windows and macOS updates so the distrusted roots are removed, including for curl and git. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Attackers exploit Apache Axis flaw CVE-2026-90436 in Confluence Data Center</title>
<link>https://xanderhutch.github.io/CyberView/#e=atlassian-confluence-axis-cve-2026-90436-2026-11</link>
<guid isPermaLink="false">cyberview:atlassian-confluence-axis-cve-2026-90436-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 23:02:00 GMT</pubDate>
<category>vuln</category>
<description>Rapid7 says its MDR team detected exploitation of CVE-2026-90436 against a customer&apos;s Confluence Data Center on November 13. The flaw is unauthenticated remote code execution in end-of-life Apache Axis 1.x, reachable when Confluence&apos;s Remote API is enabled. The intruder harvested database credentials, queried a cloud data warehouse and later signed in to Microsoft 365 as a service account. Confluence 9.2.28 fixes it; Cloud is unaffected.

What to do: Upgrade Confluence Data Center to 9.2.28 or a current 10.x release, internet-facing first. If you cannot, switch off Remote API (XML-RPC &amp; SOAP). Then check exposed servers for compromise with Rapid7&apos;s indicators.

CVEs: CVE-2026-90436 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Openwave Webtop Media Transcoder: file-type restriction bypass added to CISA exploited list</title>
<link>https://xanderhutch.github.io/CyberView/#e=kev-cve-2026-111793</link>
<guid isPermaLink="false">cyberview:kev-cve-2026-111793</guid>
<pubDate>Tue, 17 Nov 2026 22:59:37 GMT</pubDate>
<category>vuln</category>
<description>The Webtop Media Transcoder does not enforce the file-type limits an administrator sets, so a signed-in user who goes around the Webmail interface can have it process and convert document types that were meant to be blocked. CISA lists the flaw as exploited and set a federal due date of 2026-11-24.

What to do: Apply the vendor&apos;s mitigations, check each affected system&apos;s internet exposure and follow CISA&apos;s BOD 26-04 triage guidance; stop using the transcoder if no mitigation is available. Federal due date is November 24.

CVEs: CVE-2026-111793 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>SignDrop: researcher releases Windows zero-day giving SYSTEM privileges</title>
<link>https://xanderhutch.github.io/CyberView/#e=windows-signdrop-zero-day-2026-11</link>
<guid isPermaLink="false">cyberview:windows-signdrop-zero-day-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>vuln</category>
<description>A researcher using the name MSNightmare published SignDrop on November 10, an unpatched Windows flaw that lifts a standard user to SYSTEM by abusing CLFS log authentication and Microsoft Defender&apos;s signature-update interface. Cyderes reproduced it in a controlled environment and got a SYSTEM shell. There is no CVE or patch; as of November 11 Microsoft had not commented, and no use in real attacks had been confirmed.

What to do: No patch exists. Limit who can run untrusted software with application control, do not disable CLFS authentication, deploy Cyderes&apos;s detections for the Defender update interface and unexpected hid.dll files, and apply Microsoft&apos;s fix when released. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Unpatched Check Point gateways wiped in PHP2Shell attacks</title>
<link>https://xanderhutch.github.io/CyberView/#e=check-point-php2shell-gateway-wiper-campaign-2026-11</link>
<guid isPermaLink="false">cyberview:check-point-php2shell-gateway-wiper-campaign-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>incident</category>
<description>Check Point says attackers chained the PHP2Shell flaws in the Mobile Access Portal with a Gaia privilege escalation to gain root on unpatched gateways from October 30. A first wave hit Israeli organisations and wiped some systems, including a VMware ESXi host; a second wave from November 2 exploited gateways in several countries without destruction. About two dozen unpatched systems were compromised.

What to do: Install Check Point&apos;s November Jumbo Hotfix on all Management Servers and Security Gateways, starting with internet-facing Mobile Access deployments. Hunt for the published indicators and ask Check Point Support for an exposure assessment.

CVEs: CVE-2026-109618, CVE-2026-109619, CVE-2026-123039 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>German military grounds Quantum Systems drones after cyberattack on the maker</title>
<link>https://xanderhutch.github.io/CyberView/#e=quantum-systems-drone-maker-cyberattack-2026-11</link>
<guid isPermaLink="false">cyberview:quantum-systems-drone-maker-cyberattack-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>incident</category>
<description>Quantum Systems confirmed an active cyberattack by &quot;hostile actors&quot;, noticed in October, that it says its teams contained. Der Spiegel and the Financial Times, as relayed by Security Affairs and Euronews, report attackers may have used a malicious update and that the German military stopped using its drones. On 13 November the company told Reuters the Bundeswehr drones were flying again. No attacker is named.

What to do: Quantum Systems operators: follow the company&apos;s and your authority&apos;s guidance; the company says there is no risk in operating its systems. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Cursor says stolen Fireworks AI key led to trojanized Claude Code extension</title>
<link>https://xanderhutch.github.io/CyberView/#e=cursor-fireworks-ai-supply-chain-2026-11</link>
<guid isPermaLink="false">cyberview:cursor-fireworks-ai-supply-chain-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>incident</category>
<description>Cursor&apos;s November 13 update says an access key it had issued to Fireworks AI was exposed by that supplier and used to copy coding agent traces from a read-only bucket. Credentials found in the traces opened further internal tools. On October 29 the attacker placed malicious Claude Code extension builds 2.1.312 and 2.1.313 in Cursor&apos;s marketplace cache for about 11 hours; at most 5,000 users could have received them.

What to do: If you installed or updated the Claude Code extension to 2.1.312 or 2.1.313 from Cursor&apos;s marketplace on October 29, follow Cursor&apos;s October 30 notice (uninstalling alone may not be enough) and rotate any keys Cursor or Fireworks flagged. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Huntress: rogue ScreenConnect clients pushed to hundreds of hosts in SCKit attacks; ConnectWise disables extensions</title>
<link>https://xanderhutch.github.io/CyberView/#e=connectwise-screenconnect-sckit-attacks-2026-11</link>
<guid isPermaLink="false">cyberview:connectwise-screenconnect-sckit-attacks-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>vuln</category>
<description>Huntress reports two waves of attacks, on November 11 and 13, using ScreenConnect&apos;s Run Command feature to install rogue clients on 288 and then 205 hosts, apparently via a flawed third-party RMM+ extension. The SCKit toolkit tries to steal a technician&apos;s session cookie and replay it against the panel; Huntress has not seen that succeed. It names CVE-2026-84869 (fixed in 26.6.5) only as background, not as SCKit&apos;s method.

What to do: Confirm technician clients run ScreenConnect 26.6.5 or later, disable RMM+ extension 1.0.31 and 1.0.32, review Run Command history, and check endpoints for a second ScreenConnect client. Do not connect to suspect hosts through ScreenConnect.

CVEs: CVE-2026-84869 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Amazon uncovers AI agent campaign hunting exposed apps for AI provider keys</title>
<link>https://xanderhutch.github.io/CyberView/#e=amazon-ai-agent-campaign-litellm-dify-keys-2026-11</link>
<guid isPermaLink="false">cyberview:amazon-ai-agent-campaign-litellm-dify-keys-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>report</category>
<description>Amazon observed two campaigns between October 3 and November 5 that it assesses, with moderate confidence, were run with ARTEX AI agents by Chinese-nexus groups. One hunted internet-exposed AI gateways and app builders for upstream provider keys; the other tried to beat identity checks at two crypto exchanges with real-time deepfake video and forged ID photos. One vendor&apos;s server rejected every deepfake attempt the agents recorded.

What to do: Remove default or hardcoded credentials (including the LiteLLM default master key), require login on AI gateways and builders, patch Langflow and RAGFlow, rotate keys ever exposed in code, and watch for unexpected AI spend or usage.

CVEs: CVE-2025-3248, CVE-2026-28797 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Breach at La Poste subsidiary Maileva exposes 22 million people in France</title>
<link>https://xanderhutch.github.io/CyberView/#e=la-poste-maileva-breach-2026-11</link>
<guid isPermaLink="false">cyberview:la-poste-maileva-breach-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>incident</category>
<description>A cyberattack on Maileva, La Poste&apos;s mail-services subsidiary, exposed first names, last names and postal addresses of 22 million people (next.ink: potentially exposed), mostly recipients of mail sent for client organisations. La Poste says the regulator CNIL was notified and a complaint filed; next.ink reports the Paris prosecutor opened an investigation. Maileva says bank details were not compromised. The attack dates from mid-September.

What to do: Organisations using Maileva: ask which of your recipients are affected and whether you must notify them or the regulator. People in France: be wary of letters, calls or messages that use your name and address to seem genuine. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>N-able patches 17 more N-central flaws as attackers exploit the platform and deploy crypto miners</title>
<link>https://xanderhutch.github.io/CyberView/#e=n-able-n-central-zero-day-exploitation-2026-11</link>
<guid isPermaLink="false">cyberview:n-able-n-central-zero-day-exploitation-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>vuln</category>
<description>N-able says it has been responding since November 9 to N-central zero-days published without coordinated disclosure. Hotfix 2 fixed seven internet-facing flaws; Hotfix 3 fixes 17 more, CVE-2026-129271 to CVE-2026-129287, and permanently removes the Legacy API. The Hacker News reports crypto miners deployed on vulnerable servers, and Rapid7 says N-able reported exploitation on November 13.

What to do: Install N-central 2026.4 Hotfix 3 now and confirm the Legacy API is off. Upgrade all Agents and Probes following N-able&apos;s guidance. Check N-central servers for unexpected processes or high processor use. If an IT provider manages your systems, ask them to confirm.

CVEs: CVE-2026-90436, CVE-2026-129271, CVE-2026-129272, CVE-2026-129273, CVE-2026-129274, CVE-2026-129275, CVE-2026-129276, CVE-2026-129277, CVE-2026-129278, CVE-2026-129279, CVE-2026-129280, CVE-2026-129281, CVE-2026-129282, CVE-2026-129283, CVE-2026-129284, CVE-2026-129285, CVE-2026-129286, CVE-2026-129287 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Six unpatched Danfoss refrigeration controller flaws allow remote takeover</title>
<link>https://xanderhutch.github.io/CyberView/#e=danfoss-refrigeration-controller-flaws-2026-11</link>
<guid isPermaLink="false">cyberview:danfoss-refrigeration-controller-flaws-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:50:00 GMT</pubDate>
<category>vuln</category>
<description>Midnight Blue found six pre-authentication remote code execution flaws in Danfoss AK-SM800A refrigeration controllers, affecting firmware up to and including 5.3.2. CVE numbers are pending and details are withheld during coordinated disclosure. The researchers say patching alone will not make these devices safe and that hundreds remain reachable from the internet. Attack waves on such controllers began in France on October 11.

What to do: List every AK-SM800A reachable from the internet, including through integrators&apos; 4G modems, and restrict management access to a VPN, private APN or strict IP whitelist. Put controllers on a separate OT network and change default passwords. Do not rely on patching alone. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Cyberattack closes emergency room at Greiz-Ronneburg hospital in Germany</title>
<link>https://xanderhutch.github.io/CyberView/#e=greiz-ronneburg-hospital-cyberattack-2026-11</link>
<guid isPermaLink="false">cyberview:greiz-ronneburg-hospital-cyberattack-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:36:00 GMT</pubDate>
<category>incident</category>
<description>The Greiz district says a cyberattack on its hospital was discovered on Sunday, November 15, made the emergency room unavailable, and that the state criminal police and prosecutors were called in. The district calls it an extortion attempt, Kommunal.de reports; the clinic gives no information on patient data. Freie Presse reported on November 17 that the emergency room is taking patients again; Security Affairs, citing MDR, says police are investigating a suspect.

What to do: General advice: patients should check with the hospital about any postponed appointments. Other hospitals should confirm they can run emergency care without IT and that backups are kept offline. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Microsoft 365 outage knocks out Outlook, Teams and Copilot in 74 countries</title>
<link>https://xanderhutch.github.io/CyberView/#e=microsoft-365-global-outage-2026-11</link>
<guid isPermaLink="false">cyberview:microsoft-365-global-outage-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:28:00 GMT</pubDate>
<category>outage</category>
<description>Outlook, Teams and Copilot were disrupted from 12 November in 74 countries, per Al Jazeera; CyberCube puts the broader incident at about 38 hours. Microsoft first cited a network capacity upgrade, not malicious activity (via Beinsure). Its post-incident report, quoted by KIRO 7, says its testing &quot;did not adequately represent the combination of hardware and traffic mix&quot;; KIRO 7 calls the cause a months-old software defect. Acrisure estimates insured losses could reach $700 mn.

What to do: No security action is needed; Microsoft says this was not an attack. Microsoft 365 customers: confirm that delayed email has arrived, check for service credits under your agreement and review how staff communicate when email is down. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>JPMorgan Chase outage delays payments; Chase said software bug, later vendor breach unconnected</title>
<link>https://xanderhutch.github.io/CyberView/#e=jpmorgan-chase-banking-outage-2026-11</link>
<guid isPermaLink="false">cyberview:jpmorgan-chase-banking-outage-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:28:00 GMT</pubDate>
<category>outage</category>
<description>Chase customers had trouble with online banking, transactions and Zelle payments for hours on November 16. Chase said that day it was a software bug in on-premise hardware, with no evidence it was cyber-related; the FBI told NewsNation it saw no malicious activity. On November 19 Chase said on X that vendors were hit by a cyber attack, with no connection to the disruption; KTEN&apos;s headline wrongly links the two.

What to do: Chase customers: check that November 16 transfers and Zelle payments went through once, and watch for phishing using the incident. Defenders: see the linked JPMorgan vendor event. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Google warns fully automated AI attacks are arriving</title>
<link>https://xanderhutch.github.io/CyberView/#e=google-autonomous-attacks-warning-2026-11</link>
<guid isPermaLink="false">cyberview:google-autonomous-attacks-warning-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:28:00 GMT</pubDate>
<category>report</category>
<description>Google’s November 16 threat report says that, for the first time, it found actors using LLM frameworks to automate vulnerability discovery and exploitation with no human in the loop. Code comments are in Russian and server telemetry suggests interest in French media, entertainment and education sites, but Google names no group or state. Other actors are testing agentic tools, and Google expects more autonomy over time.

What to do: Shorten patch windows for internet-facing systems, check that monitoring can respond to fast automated activity, and follow the recommendations in Google&apos;s report. Patch PaperCut servers and LiteLLM gateways (CVE-2026-40217).

CVEs: CVE-2026-81578, CVE-2026-82078, CVE-2026-40217 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>NSA&apos;s classified internal network NSA Net goes down</title>
<link>https://xanderhutch.github.io/CyberView/#e=nsa-net-classified-network-outage-2026-11</link>
<guid isPermaLink="false">cyberview:nsa-net-classified-network-outage-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:28:00 GMT</pubDate>
<category>outage</category>
<description>CNN reported, citing six people, that NSA Net went down on 17 November; other sources say service was intermittent. The NSA cites degradation during scheduled routine maintenance; DNI Jay Clayton linked it to PHP patching. CNN&apos;s sources said there was no evidence of a cyberattack. On 20 November an NSA spokesperson told Nextgov: &quot;NSA can confirm that we are now operating under normal conditions.&quot;

What to do: No action needed; for awareness. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Equinor cyberattack on IT systems stopped, company says; production unaffected</title>
<link>https://xanderhutch.github.io/CyberView/#e=equinor-ongoing-cyberattack-2026-11</link>
<guid isPermaLink="false">cyberview:equinor-ongoing-cyberattack-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:28:00 GMT</pubDate>
<category>incident</category>
<description>Equinor said on its website on 17 November it had detected an ongoing cyberattack on certain IT systems. On 19 November it said the attack did not appear to target Equinor specifically. Its 20 November update says the attack has been stopped, production is running as normal, there is no indication data was stolen or altered, and the matter has been reported to the police.

What to do: Equinor partners and suppliers: no action required yet. Follow Equinor&apos;s official updates, treat unexpected emails that appear to come from Equinor with extra care, and confirm payment or access requests by phone. (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Chrome 157 fixes eighth exploited zero-day of 2026</title>
<link>https://xanderhutch.github.io/CyberView/#e=chrome-157-zero-day-cve-2026-125350</link>
<guid isPermaLink="false">cyberview:chrome-157-zero-day-cve-2026-125350</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>vuln</category>
<description>Google&apos;s November 9 stable update takes Chrome to 157.0.8103.50/.51 with nine security fixes. Google says an exploit exists in the wild for CVE-2026-125350, a high-severity use-after-free in XML. Ivanti counts it as Chrome&apos;s eighth zero-day of 2026 (via CSO Online). CISA has added it to its Known Exploited Vulnerabilities catalog.

What to do: Update Google Chrome to 157.0.8103.50 or later and relaunch the browser; do not wait for the staged rollout. IT teams: push the update to managed devices and check that other Chromium-based browsers have their equivalent fix.

CVEs: CVE-2026-125350, CVE-2026-125344, CVE-2026-125343, CVE-2026-125345, CVE-2026-125347, CVE-2026-125348 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>SAP Patch Day: 35 notes, four rated CVSS 10.0</title>
<link>https://xanderhutch.github.io/CyberView/#e=sap-patch-day-2026-11</link>
<guid isPermaLink="false">cyberview:sap-patch-day-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>vuln</category>
<description>SAP&apos;s November 10 CVE records confirm four flaws rated CVSS 10.0: memory corruption in the NetWeaver AS ABAP Internet Communication Manager and in Web Dispatcher, a SAP Gateway default access list that can treat outsiders as internal servers, and a SAML2 assertion check failure allowing impersonation of any user. Each is fixed by its own SAP Note. The 35-note total comes from secondary coverage.

What to do: SAP administrators: apply SAP Notes 3778530, 3778556, 3798345 and 3849308 as a priority. All four flaws are rated 10.0 and need no login. Cover every Web Dispatcher and Gateway instance and any AS ABAP system using SAML2 logon.

CVEs: CVE-2026-76964, CVE-2026-89126, CVE-2026-89103, CVE-2026-120017 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>React Server Components flaw allows unauthenticated remote code execution</title>
<link>https://xanderhutch.github.io/CyberView/#e=react-server-components-cve-2026-126187</link>
<guid isPermaLink="false">cyberview:react-server-components-cve-2026-126187</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>vuln</category>
<description>CVE-2026-126187 is a deserialisation-of-untrusted-data flaw in React Server Components that allows remote code execution without a login. It affects the react-server-dom-webpack, parcel and turbopack packages in the 19.0, 19.1, 19.2 and 19.3 lines; 19.0.9, 19.1.10, 19.2.9 and 19.3.1 fix it. Canada&apos;s Cyber Centre and South Korea&apos;s KISA both issued notices on November 12. Neither reports exploitation.

What to do: Check for react-server-dom-webpack, react-server-dom-parcel or react-server-dom-turbopack in your applications and update to 19.0.9, 19.1.10, 19.2.9 or 19.3.1, matching your release line, then redeploy. Ask web application suppliers to confirm.

CVEs: CVE-2026-126187 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Microsoft Patch Tuesday: 1,093 fixes and an exploited Outlook zero-day</title>
<link>https://xanderhutch.github.io/CyberView/#e=microsoft-patch-tuesday-2026-11</link>
<guid isPermaLink="false">cyberview:microsoft-patch-tuesday-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>vuln</category>
<description>The second four-digit release in a row includes 63 critical flaws. CVE-2026-124194, an Outlook spoofing bug that leaks NTLMv2 hashes when a crafted message is opened, is under active attack. Analysts also flag a Windows Search remote code execution bug rated 9.8.

What to do: Install the November 2026 Microsoft updates, starting with Outlook (CVE-2026-124194, exploited) and the 9.8-rated Windows LDAP, Windows Search and Xbox Live Networking flaws. Tell users not to open unexpected message-file attachments.

CVEs: CVE-2026-124194, CVE-2026-116928, CVE-2026-119958, CVE-2026-112967, CVE-2026-119990, CVE-2026-114160 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>CISA adds six exploited flaws to KEV: Cisco BroadWorks, Barracuda ESG, Outlook, Chromium, Openwave</title>
<link>https://xanderhutch.github.io/CyberView/#e=cisa-kev-six-additions-2026-11-10</link>
<guid isPermaLink="false">cyberview:cisa-kev-six-additions-2026-11-10</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>vuln</category>
<description>The six additions are an XML external entity flaw in Cisco BroadWorks, an authentication bypass and a command injection in Barracuda products including Email Security Gateway, an Outlook flaw (CVE-2026-124194), a Chromium use-after-free (CVE-2026-125350) and a flaw in Openwave Webtop Media Transcoder. Federal due dates in the catalog run from November 13 to November 24.

What to do: Apply vendor fixes for Cisco BroadWorks, Barracuda ESG, Outlook, Chromium browsers and Openwave Webtop, starting with internet-exposed systems, and check whether they were compromised before patching. Federal due dates: Barracuda Nov 13, others Nov 24.

CVEs: CVE-2026-20320, CVE-2026-116606, CVE-2026-111560, CVE-2026-124194, CVE-2026-125350, CVE-2026-111793 (Written by CyberView AI (a Claude model).)</description>
</item>
<item>
<title>Ymir Link: a total of 47,864,147 records leaked from 49 Cuenote FC systems, company says</title>
<link>https://xanderhutch.github.io/CyberView/#e=ymir-link-cuenote-email-records-2026-11</link>
<guid isPermaLink="false">cyberview:ymir-link-cuenote-email-records-2026-11</guid>
<pubDate>Tue, 17 Nov 2026 22:08:00 GMT</pubDate>
<category>incident</category>
<description>Ymir Link&apos;s third report, dated 11 November, says program vulnerabilities were exploited and 47,864,147 records leaked from 49 Cuenote FC systems, revised down from 51,628,280 after removing double counts; it says no payment card data is confirmed leaked. On 20 November Peach Aviation said 8,997,106 of its customers&apos; personal records leaked through Cuenote FC, Cybersecurity.com (Japan) reported.

What to do: Organisations that send email through Cuenote FC: ask Ymir Link which of your lists were taken, assess whether you must notify recipients, and warn them about fake emails. Recipients: be cautious of unexpected messages from brands you subscribe to. (Written by CyberView AI (a Claude model).)</description>
</item>
</channel>
</rss>
