ICHI Briefs & Podcasts · ICHI Daily · five-minute podcast

ICHI Global Five Minute Cyber Rundown

ICHI Daily · Published · 5 min listen

Check Point gateways chained to root with today's federal patch deadline, South Korea's government email breached through single sign-on, an AI-labelled hijack of Luxor mining infrastructure, Cursor's full incident account, and a code-signing server targeted through Log4Shell.

Listen to this episode

MP3, 5.1 MB · · the full transcript is below.

Show notes

ICHI Global Five Minute Cyber Rundown — Monday, November 16, 2026

Today's theme is trust boundaries failing in the same weekend: perimeter firewalls chained to root with a federal deadline today, a government email system reached through single sign-on, an AI-labelled hijack of mining infrastructure, and two development-pipeline intrusions — one contained, one apparently stopped before a supply-chain attack was completed.

Most important development — Check Point gateway chain, federal deadline today (Nov 16)

Check Point Research has documented exploitation of unpatched Check Point Security Gateways beginning October 30, 2026, about a week after fixes became available. Attackers gained initial access through Mobile Access portals via the PHP2Shell flaws (CVE-2026-109618 and CVE-2026-109619), then escalated to root through CVE-2026-123039, a local privilege-escalation flaw in the Gaia OS DynamicCLI interface: the root-owned dynamic_server process accepts commands over /tmp/dynamic_socket, a world-writable local socket that does not authenticate connecting processes. The flaw therefore requires an existing local foothold — it is not a standalone unauthenticated remote takeover — which is why the portal-to-root chain matters.

After root, attackers deployed PHP webshells and the persistent TSGA backdoor with its embedded TSProxy component (variants of the PhiliKit framework previously associated with Ivanti exploitation), scanned internally, and in the first wave pivoted over SSH to a VMware ESXi host and ran the destroy_v6 wiper, leaving politically themed DREAM_ANNEXATION.txt notes. The first wave (Oct 30) targeted Israeli organizations and included destructive activity against some of them; a broader second wave (Nov 2) targeted organizations across multiple countries and sectors with no evidence of destructive activity. Check Point says roughly two dozen customers were compromised. Attribution is not assessed with high confidence: the tooling lineage points toward a China-nexus association, but the destructive activity and targeting do not fully align, and Check Point lists initial-access-broker and false-flag scenarios as plausible.

CISA added CVE-2026-123039 to its Known Exploited Vulnerabilities catalog on November 13, 2026 with a federal remediation deadline of today, November 16, 2026. Fixes are in the November Jumbo Hotfix takes (including R82.20 Take 14, R82.10 Take 52, R82 Take 132, and R81.20 Take 174). Spark Firewalls and Smart-1 Cloud are unaffected.

Defence: install the November Jumbo Hotfix on every affected gateway, management server and log server today, prioritizing internet-facing Mobile Access deployments — then hunt for compromise using Check Point's published indicators (zz*.tpl webshells, slot_CPconfigdaemon / TSGA, TSProxy) before assuming a patched box is clean. Patching does not remove an implant already present.

Confirmed incident — South Korea government email, ~300 accounts (announced Nov 16)

South Korea's Ministry of the Interior and Safety said on Monday, November 16, 2026 that hackers infiltrated the government's official email system through vulnerable local government networks, exploiting a flaw in the single sign-on authentication system to gain unauthorized access to some 300 accounts. The ministry reported the breach to the privacy watchdog, notified affected users, blocked unnecessary connections to the system, and launched a joint investigation with the National Intelligence Service and the Korea Internet & Security Agency. No attribution or data-exfiltration finding has been published.

Defence: treat single sign-on as a tier-zero boundary — the weakest network that can reach it sets its effective security. Audit which networks and systems can reach government and enterprise SSO, and review SSO logs for anomalous authentications.

AI + cyber watch — Luxor Commander hijack (Sat Nov 14; company statement Nov 15)

Bitcoin mining services firm Luxor says attackers compromised its Commander mining-management tool on Saturday (CEO post: about 6:00 PM PDT) through what CEO Nick Hansen called a "sophisticated AI-enabled vector." The attackers redirected customers' mining power to the OCEAN pool with payouts assigned to an attacker-controlled address. Luxor says it detected the unauthorized changes, patched the tool within two hours, restored affected machines to their original pool settings, and will compensate affected customers in full; it says the incident was limited to Commander — Luxor Pool and LuxOS were not affected — and that no data was leaked or exfiltrated. Luxor puts the scope at just over a dozen customer accounts.

Unconfirmed: early social-media reports claimed about 64,000 machines / 13 EH/s were hijacked; Luxor has not confirmed either number. Blockchain researchers cited in coverage linked an address that received two OCEAN payouts worth about 0.162 BTC (~$10,500). Critically, Luxor has not explained how AI was used in the attack; the "AI-enabled" characterization is the company's description, not a verified method. A post-mortem is promised. This follows October's Chainalysis disclosure of an attacker using "offensive agentic techniques."

Defence: mining-management and other fleet-management planes are tier-zero — restrict and monitor who can change pool/destination configuration, and alert on destination changes.

Confirmed incident — Cursor post-mortem: vendor key to malicious extension (published Nov 13)

Cursor published its full incident update (post dated November 13, 2026). The chain began at vendor Fireworks: an attacker obtained an access key Cursor had issued to Fireworks, which Fireworks had exposed, and used it to copy data from a read-only storage bucket containing limited coding-agent traces (mostly from 2025, from sessions not in Privacy Mode). A small number of traces contained unredacted credentials; fewer than 0.01% of customers had API credentials exposed, and traces were exposed for fewer than 0.05% of users. Using credentials from that data, the attacker reached select SaaS tools, archived non-production source repositories, limited support data, usage metrics and conversation summaries.

On October 29, using a stolen credential, the attacker placed malicious, unauthorized copies of Anthropic's Claude Code extension (versions 2.1.312 and 2.1.313) directly on the cache serving Cursor's VS Code extension marketplace. At most 5,000 users who requested those cached versions potentially received the malicious build; it was available for about 11 hours before removal. Anthropic did not publish the copy, and Open VSX never had the malware. Cursor says production systems, customer code repositories, release signing keys and the Cursor app itself were not affected; the last attacker activity it saw was November 1 and was blocked, and it believes the incident is contained. Affected customers were contacted directly; uninstalling the extension alone may not be enough for those who installed the tampered versions.

Defence: rotate any credentials that ever appeared in AI coding-agent traces or support tickets; treat vendor-issued keys and trace buckets as sensitive stores, with secret scanning and redaction verified — not assumed.

Adversary activity — BO Team at a Russian developer; code-signing server targeted (Kaspersky)

Kaspersky's ICS CERT has documented a months-long compromise of an unnamed Russian software developer (the developer contacted Kaspersky in June 2025), attributed to BO Team (also Black Owl / Hoody Hyena / Lifting Zmiy). Entry was through an exposed SSH service on a Linux terminal server; the attackers then exploited Log4Shell (CVE-2021-44228) in VMware vCenter 7.0.3 to obtain root, backdoored a hypervisor with an hourly cron script disguised as a PostgreSQL component ("pgcache") using the GSocket utility, and compromised — per Kaspersky — all components of the IT infrastructure, including GitLab, Ansible, backups and domain credentials (LSASS and ntds.dit extraction).

The server of particular interest signed compiled builds of the developer's commercial software. Kaspersky assesses the attackers may have set out to inject malicious code into a released product (a supply-chain attack) or steal the digital signature, and says that after detection the attackers lost access and were apparently unable to bring the attack to its intended conclusion; a follow-on phishing campaign in late July, sent with a compromised DKIM key, was detected and prevented. No trojanized build delivered to customers has been established.

Defence: a five-year-old, widely known flaw in vCenter still yields hypervisor root — inventory and patch virtualization management planes, and isolate and monitor build-signing servers as tier-zero assets.

Defensive warning — Microsoft 365 outage phishing (Microsoft Threat Intelligence, Nov 14)

Microsoft Threat Intelligence warned on November 14, 2026 of an active phishing campaign exploiting the November 12 Microsoft 365 / Exchange Online disruption. Messages impersonate Microsoft 365 service-health notices, quote the genuine incident number EX1550543, and claim mail held during the outage will be deleted within 24 hours unless the recipient clicks a recovery link. Microsoft did the opposite during the outage — it extended email timeouts so delayed mail would not be lost. Reported indicators include subjects of the form "Outage - [six-character code]", sender domains moments-notice[.]com, planetmaison[.]com and newbooker[.]com, and recipient-specific SendGrid tracking links redirecting to reallifeaccounting[.]com. Microsoft describes the activity as consistent with credential phishing; credential theft itself is not confirmed, and Microsoft has said the outage itself was a technical issue, not an attack.

Defence: tell users now — Microsoft does not ask anyone to click a link to restore held mail; verify service status in the Microsoft 365 admin center, and hunt Defender for Office 365 / Defender XDR email and URL-click tables for the reported domains.

Defensive implications

  • Check Point: November Jumbo Hotfix today (KEV deadline Nov 16), then forensic hunt — webshells and TSGA/TSProxy survive patching.
  • Identity: audit every network path that can reach single sign-on; review SSO authentication logs (South Korea case).
  • Management planes: Commander-style fleet/mining managers and RMM tools — lock down and alert on configuration/destination changes.
  • AI tooling hygiene: rotate credentials exposed in agent traces or support tickets; verify redaction and secret scanning on vendor-shared data (Cursor case).
  • Virtualization + signing: patch vCenter-class management planes regardless of flaw age; isolate build-signing servers (BO Team case).
  • User warning: circulate the Microsoft 365 outage-phishing indicators before users click.

ICHI note

No live ICHI composite score is cited in this episode because the live score did not render in the pre-production check at production time. ICHI.global's production methodology remains VECTORS v7.1; no methodology change is announced. Model C scoring refinements remain under development.

SOURCES

  • Check Point Research — PHP2Shell Exploitation on Check Point Gateway: https://research.checkpoint.com/2026/php2shell-exploitation-on-check-point-gateway/
  • GBHackers — CISA Warns of Actively Exploited Check Point Zero-Day Vulnerability Allowing Attackers to Gain Root Access: https://gbhackers.com/cisa-check-point-zero-day-vulnerability/
  • Aju Press — Hackers breach 300 government email accounts: https://www.ajupress.com/view/20261116135007197
  • CoinCentral — Luxor Promises Full Refunds After Attackers Redirect Customers' Bitcoin Mining Power: https://coincentral.com/luxor-promises-full-refunds-after-attackers-redirect-customers-bitcoin-mining-power/
  • Cursor — An update on our recent security incident: https://cursor.com/blog/security-incident-update
  • The Hacker News — BO Team Exploits Log4Shell in VMware vCenter, Compromises Russian Firm's Code-Signing Server: https://thehackernews.com/2026/11/bo-team-exploits-log4shell-in-vmware.html
  • SANS Internet Storm Center via CyberSecPods — SANS Stormcast Monday, November 16th, 2026: https://cybersecpods.com/podcasts/sans-internet-stormcenter-daily-cyber-security-podcast-stormcast/sans-stormcast-monday-november-16th-2026-macfinger-clickfix-downthedrain-microso/
  • Vijilan Security — M365 Outage Nov 12, 2026: What to Check: https://vijilan.com/blog/microsoft-365-outage-november-12-2026

Transcript

Akari: The daily cyber rundown, brought to you by I C H I dot global. I'm Akari, an AI-generated voice. This is Monday, November sixteenth, twenty twenty-six. Over the weekend the pattern held, perimeter firewalls under active attack with a federal patch deadline that lands today, a government email system breached through single sign-on, and another A I assisted intrusion, this time against Bitcoin mining infrastructure.

Akari: The most important development is the Check Point gateway campaign, and today is the deadline. Check Point Research has documented attackers breaking into unpatched Security Gateways through the Mobile Access portal, then using a flaw in the Gaia operating system, tracked as C V E twenty twenty-six, one two three zero three nine, to climb from the portal account to full root control. From there they planted webshells and a persistent backdoor, and in the first wave, which targeted Israeli organizations on October thirtieth, they moved from a gateway into a VMware host and ran a wiper. A second, broader wave followed on November second. Check Point says roughly two dozen customers were compromised. The flaw only gives root to someone who already has a foothold, which is exactly why the chain matters. The federal remediation deadline for this flaw is today, November sixteenth, so patch every gateway and management server to the November hotfix, and then hunt for the webshells and backdoor Check Point has published indicators for, because patching alone does not remove an implant that is already there.

Akari: Next, a confirmed government breach in South Korea. The interior ministry said today that hackers reached the government's official email system through vulnerable local government networks, exploiting a flaw in the single sign-on authentication system to access about three hundred accounts. The ministry has reported the breach to the privacy watchdog, notified the affected users, blocked unnecessary connections, and opened a joint investigation with the National Intelligence Service and the Korea Internet and Security Agency. It is another reminder that single sign-on is a single point of failure when the networks feeding into it are not equally hardened.

Akari: Now the weekend's A I story, with an important caution. Luxor, a Bitcoin mining services firm, says attackers compromised its Commander mining management tool on Saturday through what its chief executive called a sophisticated A I enabled vector, redirecting the mining power of just over a dozen customer accounts to another pool. Luxor patched the tool within two hours, says no data was leaked, and has promised to repay affected customers in full. Early social media reports claimed sixty-four thousand machines were hijacked, but Luxor has not confirmed that figure, and the company has not yet explained how A I was actually used. Treat the A I label as the company's description, not a verified method, until its promised post-mortem appears.

Akari: Two development world stories deserve your attention. Cursor has published the full account of its October incident. An attacker obtained a vendor access key and copied agent session traces containing unredacted credentials, reached internal tools, and on October twenty-ninth placed a malicious copy of the Claude Code extension in Cursor's marketplace cache, where at most five thousand users may have received it over about eleven hours. Cursor says the incident is contained. And Kaspersky has documented a months-long intrusion at a Russian software developer, where attackers entered through an exposed S S H service, used the old Log4Shell flaw in VMware vCenter to reach the hypervisors, and focused on the server that signs the company's software builds. Kaspersky believes a supply chain attack was the likely goal, and says the attackers lost access before completing it.

Akari: Finally, a phishing warning. Microsoft Threat Intelligence says attackers are exploiting last week's Microsoft three sixty-five outage with fake service health emails that quote the real incident number and claim held mail will be deleted within twenty-four hours unless you click to restore it. Microsoft never extended that threat, it extended mail timeouts instead. Tell your users, do not click, check service status in the admin center. So, defensively, patch and hunt on Check Point today, audit who can reach your single sign-on and from where, rotate any credentials that ever passed through an A I coding tool, and verify your build signing servers are isolated and monitored. The takeaway is that the perimeter, the identity layer, and the build pipeline were all hit in the same weekend, and each one failed at a trust boundary. For the underlying evidence behind every story today, visit I C H I dot global. I'm Akari. Stay safe, stay frosty, and I will see you tomorrow.

Sources

  1. Check Point Research — PHP2Shell Exploitation on Check Point Gateway research.checkpoint.com
  2. GBHackers — CISA Warns of Actively Exploited Check Point Zero-Day Vulnerability Allowing Attackers to Gain Root Access gbhackers.com
  3. Aju Press — Hackers breach 300 government email accounts www.ajupress.com
  4. CoinCentral — Luxor Promises Full Refunds After Attackers Redirect Customers' Bitcoin Mining Power coincentral.com
  5. Cursor — An update on our recent security incident cursor.com
  6. The Hacker News — BO Team Exploits Log4Shell in VMware vCenter, Compromises Russian Firm's Code-Signing Server thehackernews.com
  7. SANS Internet Storm Center via CyberSecPods — SANS Stormcast Monday, November 16th, 2026 cybersecpods.com
  8. Vijilan Security — M365 Outage Nov 12, 2026: What to Check vijilan.com

Get ICHI Daily and ICHI Weekly by email

No new episode, no email. Every email links to the episode on ICHI; audio is never attached.


What to receive


✓ Daily podcast included ✓ Weekly podcast included ✓ Weekly written brief included


✓ Daily podcast included ✓ Weekly podcast included ✗ Weekly written brief not included


✗ Daily podcast not included ✗ Weekly podcast not included ✓ Weekly written brief included


✗ Daily podcast not included ✓ Weekly podcast included ✓ Weekly written brief included


✓ Daily podcast included ✗ Weekly podcast not included ✗ Weekly written brief not included


Choose any combination below, including the weekly podcast only or the weekly written brief only.

Custom: choose any combination

We send a confirmation email first, and nothing else until you confirm. Already subscribed? Use the “manage” link in any email to change what you receive; nothing changes unless you choose it. Your address is used only to send these emails; every email has a one-click unsubscribe link.

All ICHI Daily episodes · ICHI Briefs & Podcasts

ICHI Daily and ICHI Weekly are editorial commentary on ICHI’s evidence. They never change the VECTORS score, incident evidence classifications or intelligence rankings.

ICHI Briefs & Podcasts · ICHI Daily · ICHI Weekly · Episode calendar · ICHI Perspectives · episodes.json · ICHI dashboard · Methodology