CTIPilot
AI-generated · no human review · verify critical claims against the linked source. how it works →
‹Tue · 17 Nov 2026
All daily briefs →
Daily brief · UTC day

Tuesday, 17 November 2026

6 verified findings from 1 run · 2 updates to prior coverage · the settled record for this UTC day, in the classic brief order.

Criticality
Kind
Topic
Region
TL;DR · the day in one read
  1. 01OptimJack: Optimizely confirms an exploited flaw; Report URI counts 178 organisations served poisoned scripts. Optimizely says an unauthorized party exploited a vulnerability in its Experimentation product to add malicious JavaScript to some customer projects, and Report URI counts 178 organisations that served the code on 2026-11-16, among them a crypto clipper and card skimmer in the Optimizely files. Ohey reports that scripts served by Kissmetrics and StorifyMe gained a loader from 2026-11-13 that routes iPhones to the Coruna and DarkSword exploit-kit stages and that minute-ly's loader was replaced outright; Hudson Rock names 21 web properties, among them the University of Basel's StorifyMe widget, and calls the iPhone stages delivery observations, not confirmed compromises. How the Kissmetrics, StorifyMe and minute-ly files were altered is unknown. →
  2. 02Craft CMS unauthenticated SQL injection gives admin takeover; CCCS reports exploitation; update to 4.18.9 or 5.11.2. CVE-2026-123488 is an unauthenticated SQL injection in the front-end search of Craft CMS 4.0.0 to 4.18.8 and 5.0.0 to 5.11.1 on MySQL and MariaDB that gives read and write access to the database, including setting the admin flag or overwriting password hashes. The Canadian Centre for Cyber Security says open source reporting indicates exploitation in the wild and VulnCheck lists it in its own KEV database; no first-hand telemetry or CISA listing was found. Craft fixed it in 4.18.9 and 5.11.2 and offers a query-string block as a workaround. →
  3. 03GTIG quarterly: exploit_agent automates n-day exploitation, suspected DPRK cluster abuses .claude/settings.json. Google Threat Intelligence Group and Mandiant's quarterly AI threat update of 2026-11-16 reports an exposed multi-agent framework, exploit_agent, that compares code before and after patch announcements, drafts exploits and runs them with no human in the loop, the first such full automation GTIG has identified. It also reports a suspected MIDNIGHT NEPTUNE (North Korea) cluster that modified .claude/settings.json in victims' internal repositories so that Claude Code downloads the SOMBERMEME backdoor at startup, and PRC-nexus actors using an LLM platform's file-hosting backend for staging and exfiltration. →

01Active threats, incidents & disclosures4 items

HIGHupdatedNATOB2

OptimJack: poisoned scripts served by Optimizely, Kissmetrics, StorifyMe and minute-ly route iPhones to exploit-kit stages, the Optimizely files also carry a crypto clipper and card skimmer, and the University of Basel site loaded a poisoned StorifyMe widget on 2026-11-14

Per Ohey, from 2026-11-13 the scripts served by Kissmetrics and StorifyMe gained a few lines that load a small router script for iPhones, and minute-ly's loader was replaced outright (Ohey, 2026-11-17). Optimizely's customer project files followed on 2026-11-16, when Report URI saw its first injection at 04:21 UTC and counted 188 infected snippets of 178 organisations, a floor (Report URI, 2026-11-16). Optimizely says an unauthorized party exploited a vulnerability in Experimentation to add the JavaScript to some customer projects and that it has fixed the flaw, naming no flaw class or count (Optimizely, 2026-11-17). Hudson Rock recovered nine poisoned vendor files itself and names 21 web properties, among them BBC, Zoom, GitLab's marketing site, HSBC UAE and the University of Basel (Hudson Rock, 2026-11-16). The Basel capture of 2026-11-14 recorded a poisoned StorifyMe web component (Hudson Rock, 2026-11-16). How the Kissmetrics, StorifyMe and minute-ly files were altered, who operates the campaign and how many visitors loaded them are unknown (Hudson Rock, 2026-11-16).

The financial theft sits in the Optimizely files: Ohey says that on 2026-11-16 they switched to a script that swaps copied wallet addresses and skims card numbers in every browser and adds the router only on iPhones and iPads (Ohey, 2026-11-17). Hudson Rock describes the stealer as skimming card number, expiry and CVV from form and fetch POST bodies and sending readable cookies with a page image (Hudson Rock, 2026-11-16). The BBC and HSBC UAE Optimizely files also embed a full fallback copy that runs when the remote fetch fails or takes longer than 2.5 seconds, so blocking the attacker host does not stop the clipper (Hudson Rock, 2026-11-16). In Hudson Rock's reading the router sends iOS 13 to 17 to a Coruna-associated entry and iOS 18 to 25 to a DarkSword-labelled one (Hudson Rock, 2026-11-16), and Ohey states that DarkSword attacks iPhones running iOS 18.4 to 18.7.2 through Safari (Ohey, 2026-11-17); Hudson Rock calls them delivery observations, not demonstrated compromises (Hudson Rock, 2026-11-16). Report URI saw a mass clean-up of the Optimizely snippets between about 20:40 and 21:10 UTC on 2026-11-16, and a snippet that was cleaned and then re-infected shows the writer still had access at that point (Report URI, 2026-11-16).

The confirmed compromise is malicious code inside responses served from third-party providers’ own script and widget endpoints.

These are delivery observations, not demonstrated iPhone compromises.

A FAZ capture records a poisoned StorifyMe widget response. Later captures include Sydney Morning Herald, University of Basel and vidaXL.

A DNS or network block is useful, but it does not remove the poisoned trusted script.

Hudson Rock (InfoStealers) 2026-11-16

DarkSword attacks iPhones running iOS 18.4 to 18.7.2 through Safari.

GitLab's and Zoom's Optimizely files and minute-ly's loader still carried it at 21:10 UTC on 16 November.

Ohey (Kirk)

an unauthorized party exploited a vulnerability in Experimentation and used it to add malicious JavaScript to some customer projects.

Optimizely status page 2026-11-17

The real number is likely higher as we could only check snippets we could find in time.

Cards entered in hosted payment iframes (Stripe, Adyen and similar) were out of its reach.

Report URI Blog (Scott Helme) 2026-11-16
Updaterun 2026-11-18T0356Z-intelactionsbodyevidenceheadlinesourcessourcing_notesummary

Optimizely says an unauthorized party exploited a vulnerability in Experimentation and used it to add malicious JavaScript to some customer projects, that it has fixed the vulnerability and removed the injected code, and that it secured affected user accounts; the status entry names no flaw class, CVE, count or stolen credentials (Optimizely, 2026-11-17). The status entry covers the Optimizely files only; how the Kissmetrics, StorifyMe and minute-ly scripts were reached is still unknown.

Report URI's timeline adds that the card skimmer was added to the clipper at 13:43 UTC on 2026-11-16, that a mass clean-up ran between about 20:40 and 21:10 UTC, and that a snippet cleaned and then re-infected shows the writer still had access at that point (Report URI, 2026-11-16). Contradiction on dates: Ohey puts the change to the Optimizely project files at 2026-11-13 together with the Kissmetrics and StorifyMe scripts (Ohey, 2026-11-17), while Report URI's first Optimizely injection is at 04:21 UTC on 2026-11-16; Report URI's logged first-hand record is the better dated one, and the earlier Kissmetrics, StorifyMe and minute-ly wave is not in the 178 (Report URI, 2026-11-16).

The injected loader came in three forms: a plain script tag, the same address hidden in base64 and decoded with atob(), which a search for the host name misses, and a full inline copy that runs through eval (Report URI, 2026-11-16). Cards entered in hosted payment iframes were out of the skimmer's reach, while cards typed into a site's own form fields were not (Report URI, 2026-11-16).

Builds on: DeadTag: expired tag domains send visitors to iPhone exploit kits; opening the page is enough… · APT29-linked watering hole uses n-day iOS and Chrome chains; CISA lists CVE-2024-27834; updated…

threat17 Nov 04:38Zmulti-sourceOpen finding →
NOTABLENATOB2

TGR-STA-0049: Unit 42 describes the PerfectPioneer and KUKURAT implants, which use HTTP/3 over QUIC for command and control against government networks entered through unpatched Exchange servers, then spread through domain logon scripts and grant FullAccess on every mailbox

Unit 42 describes a campaign by TGR-STA-0049, also tracked as REF7707, Earth Alux and Ink Dragon, against Southeast Asian government entities since at least July 2025, and says the group's targeting also includes government organizations in South America and Europe (Unit 42, 2026-11-16). Server exploitation is the primary entry: unpatched Microsoft Exchange and web servers receive web shells, among them variants of ASPXSpy, Godzilla, Bysunue.aspx and Suo5, while phishing with decoy PDFs from lookalike government domains is secondary (Unit 42, 2026-11-16). The implants are two previously undocumented families: the Go-based PerfectPioneer RAT, active since July 2025 and run through DLL side-loading of a modified legitimate executable, decoy installers or phishing, and its modular C++ successor KUKURAT, deployed since March 2026, which loads at least 12 DLL modules from the command server into memory (Unit 42, 2026-11-16). Unit 42 says both families use HTTP/3 over QUIC as the primary command channel, with an automatic HTTP/2 fallback in the PerfectPioneer variant it analyzes, separate HTTP/2 and HTTP/3 builds of KUKURAT, and two PerfectPioneer variants, RatAAkk-tt and PerfectRat, that rely on HTTP/2 alone, and that some firewalls and inspection engines handle QUIC less well than TCP traffic (Unit 42, 2026-11-16). Two of the six PerfectPioneer variants, Pioneer and rat_http3_cc, carry a built-in blocklist that stops operators running noisy or destructive commands, and KUKURAT generates synthetic network, registry and file activity as camouflage (Unit 42, 2026-11-16).

After taking a domain controller, the operators installed malicious domain logon scripts, so each user's logon pulled KUKURAT; they also used PsExec, WMI and WinRM (Unit 42, 2026-11-16). Credentials came from LSASS dumps taken with rundll32 calling comsvcs.dll by ordinal rather than by the MiniDump name, from DCSync, and from Mimikatz, ProcDump, SharpDPAPI and SAM and SYSTEM hive dumps (Unit 42, 2026-11-16). On Exchange, the operators first listed accounts holding the ApplicationImpersonation role and, when none was found, ran a hidden PowerShell loop that granted a compromised account FullAccess on every mailbox (Unit 42, 2026-11-16). The exfiltration targeted pdf, docx, xlsx and png files whose names suggest international relations and internal political affairs (Unit 42, 2026-11-16).

the exploitation of vulnerable, unpatched Microsoft Exchange servers and web servers to deploy web shells

PerfectPioneer RAT relies on HTTP/3 as its primary C2 communication mechanism, automatically falling back to HTTP/2 in the event of connection failure.

Instead of calling the MiniDump function by name, the attackers invoked it via its ordinal hidden index number, #+000024.

Palo Alto Networks Unit 42 2026-11-16
threat17 Nov 04:40Zsingle-sourceOpen finding →
ROUTINENATOB2

BundID: automated mass login attempts against Germany's federal citizen accounts lock accounts; the BSI confirms the attack, password sign-in was switched off on 16 November and no compromise is evidenced

Germany's BSI confirmed to heise security and Handelsblatt on 2026-11-16 an IT security incident involving BundID accounts: automated, mass login attempts whose failures locked accounts, triggered automated mails and prompted account holders to reset their passwords (heise, 2026-11-16). The federal government switched off username-and-password sign-in on Monday morning for security reasons (Handelsblatt, 2026-11-16), while sign-in with the online ID card (eID) and the Elster certificate stayed possible (WEB.DE, 2026-11-16). The BSI says countermeasures it does not name repelled the attack and that it has no indication of a compromise of any account, and heise reports that locked accounts are apparently not being unlocked in bulk, so account holders must pass the password-reset flow or use another sign-in method (heise, 2026-11-16). Handelsblatt, citing government circles, describes the attack as credential stuffing, the automated replay of credentials from earlier leaks at other services (Handelsblatt, 2026-11-16). An aggregator reports that password sign-in returned the same afternoon behind a mandatory e-mail one-time code (ad-hoc-news.de, 2026-11-16); no primary source read confirms it. The same aggregator attributes the credential-stuffing label to the BSI, which neither heise nor Handelsblatt does (ad-hoc-news.de, 2026-11-16). No Swiss portal is reported affected.

Automated, mass login attempts have occurred against BundID accounts. (translated from German)

The BSI currently has no indications of a compromise of BundID accounts. (translated from German)

heise online (quoting the BSI)

According to current findings this is so-called credential stuffing, the Handelsblatt learned from government circles. (translated from German)

Handelsblatt 2026-11-16
incident17 Nov 04:43Zmulti-sourceOpen finding →
NOTABLENATOB2

Symantec: a September 2026 intrusion at a US retailer shows attackers switching to AI-generated PowerShell and, about ten minutes after first querying SolarWinds Orion, using the RabbitMQ Erlang cookie bundled with it to run a command on a second server

Symantec's Threat Hunter Team describes a September 2026 intrusion at a US retail organization in which the attackers first worked by hand on a compromised web server, with typing errors and every staged payload terminated by security software, and then, the next day, submitted about 50 long, error-free PowerShell scripts over roughly two hours, with labelled output sections and comments that read as a model's reasoning (Symantec, 2026-11-12). Symantec assesses AI assistance with high confidence and an operator relaying output to an AI tool, not an autonomous agent, with moderate confidence; it could not identify the model (Symantec, 2026-11-12). The initial infection vector is unknown (Symantec, 2026-11-12).

The earliest activity, on 2026-09-18, ran cmd.exe directly under sqlservr.exe on three SQL Servers, which Symantec reads as commands issued through the database service, most likely xp_cmdshell; on one host the attackers compiled the public EfsPotato exploit with csc.exe from base64 text decoded by certutil and saved the SAM, SYSTEM and SECURITY hives (Symantec, 2026-11-12). On a web server compromised through w3wp.exe they side-loaded payloads with a legitimate Bitdefender executable (bdreinit.exe with log.dll) and a G DATA executable (ask.exe with ask.dll) from ProgramData, used a utility called BreakChain to relaunch them so they no longer appeared as children of w3wp.exe, copied one to a file named w3wp.exe, and planted ASP web shells in the application's images folder (Symantec, 2026-11-12). The AI-style scripts then harvested credentials from connection strings, FileZilla, WinSCP and PuTTY settings, saved RDP files and IIS configuration, read the Erlang cookie file of the RabbitMQ broker that ships with Orion; running Orion's bundled erl.exe with that cookie, they called the broker on a second web server to run an encoded operating system command (Symantec, 2026-11-12). Later activity searched a probable domain controller for NTDS.dit, SYSVOL cpassword values and LAPS passwords and ran a Chisel tunnelling tool and a password sprayer (Symantec, 2026-11-12).

Symantec says no second-phase technique was new and suggests a possible link, through a shared address and the retail focus, to the operation Gambit Security described on 2026-09-22 (Symantec, 2026-11-12). No Swiss or public-sector victim is named; the case matters for the transferable chain and its AI-assisted tempo.

Triage: long, labelled PowerShell is not the signal on its own, because administrators write such scripts too; a parent of sqlservr.exe or w3wp.exe rather than an administrator session, credential-store targets and a blocked loader returning rebuilt are.

This is one of dozens of AI-assisted attacks that Symantec has investigated since May 2026.

About 10 minutes after first querying the victim's SolarWinds Orion monitoring software, the attackers had used a component bundled with it to run commands on a second server.

Symantec assesses the AI assistance with high confidence and the operator-in-the-loop model with moderate confidence.

Symantec Threat Hunter Team (Broadcom) 2026-11-12

Builds on: Three off-the-shelf AI agents ran almost the entire card-theft campaign, from discovery to…

threat17 Nov 04:39Zsingle-sourceOpen finding →
NOTABLECVE-2026-123488exploitedNATOA2

CVE-2026-123488, Craft CMS 4.0.0 to 4.18.8 and 5.0.0 to 5.11.1: an unauthenticated SQL injection in front-end search lets a visitor become an administrator on MySQL and MariaDB; the Canadian Cyber Centre reports exploitation in the wild and 4.18.9 and 5.11.2 fix it (CVSS 4.0 9.3)

Craft CMS 4.0.0 before 4.18.9 and 5.0.0 before 5.11.2 contain an unauthenticated SQL injection in front-end search: the attribute parameter is concatenated into the query unsanitized and emulated PDO prepares allow stacked statements, so a remote attacker can read and write the database, setting the admin flag or overwriting password hashes in the users table to gain administrator access (CVSS 4.0 9.3) (VulnCheck, 2026-11-06). Craft's own advisory, published on 2026-10-22 as critical, places the flaw on the front end of any site that uses craft.search, says it affects only MySQL and MariaDB, and says the write primitive was used to set the admin flag on user ID 1 (Craft CMS, 2026-10-22). The fixed releases are 4.18.9 and 5.11.2, and Craft's temporary workaround is to block front-end requests that carry the bracketed attribute token, plain or encoded, in the query string (Craft CMS, 2026-10-22). ENISA's record for CVE-2026-123488 lists Craft's advisory among its references (ENISA EUVD, 2026-11-06).

On exploitation, the Canadian Centre for Cyber Security's update of 2026-11-12 says open source reporting indicates the flaw is exploited in the wild (CCCS, 2026-11-12), and VulnCheck's advisory page states that it is in the VulnCheck KEV database (VulnCheck, 2026-11-06). No first-hand telemetry, victim statement or CISA listing was found.

Craft CMS 4.0.0 before 4.18.9 and 5.0.0 before 5.11.2 contains an unauthenticated SQL injection vulnerability in the element search endpoint because the attribute parameter is concatenated into the query unsanitized and emulated PDO prepares allow stacked statements.

This advisory is in the VulnCheck KEV database

VulnCheck (CNA advisory) 2026-11-06

Open source reporting indicates that CVE-2026-123488 is exploited in the wild.

Canadian Centre for Cyber Security (AV26-1086, Update 1) 2026-11-12

The write primitive was used to set the admin flag on user ID 1 in the users table

Craft CMS (GHSA-qcvp-ggx3-97r6) 2026-10-22
vulnerability17 Nov 04:44Zmulti-sourceOpen finding →

03Research, reports & policy1 item

NOTABLENATOB2

GTIG AI Threat Tracker, November 2026: an exposed exploit_agent framework automates patch diffing, exploit drafting and execution without a human, a suspected North Korean cluster plants a Claude Code startup download in internal repositories, and PRC-nexus actors stage and exfiltrate through an LLM platform's file hosting

Three findings from GTIG's quarterly update on adversary use of AI, published 2026-11-16, change what defenders can look for (GTIG, 2026-11-16). First, GTIG identified an exposed staging server hosting a multi-agent vulnerability research and weaponization framework, dubbed exploit_agent, that integrates frontier and local LLMs and runs continuous differential analysis of popular web application software: it compares code before and after patch announcements, analyzes the findings, drafts exploits and executes them automatically, which GTIG says is the first time it has seen LLM frameworks fully automate discovery and exploitation without a human in the loop (GTIG, 2026-11-16). The server also held scripts for internal reconnaissance, brute-forcing, installing web shells and removing other attackers' web shells, and its victim telemetry suggested interest in French media, entertainment and education websites (GTIG, 2026-11-16).

Second, in intrusions against cryptocurrency organizations in North America and Europe, a suspected MIDNIGHT NEPTUNE cluster used fraudulent recruiter profiles, sent targets to fake meeting platforms with ClickFix lures that ran likely LLM-authored staging scripts, and then pivoted into internal code repositories and modified .claude/settings.json so that Claude Code startup triggered the download of the SOMBERMEME backdoor; it masked lateral movement with GitHub personal access tokens named after Microsoft Copilot (GTIG, 2026-11-16). Third, in what GTIG calls a new trend, PRC-nexus actors abused LLM platforms and file-hosting backends for staging, dead-drop command channels and exfiltration, in one case staging lure archives and exfiltrating stolen data through a frontier model's file-hosting infrastructure (GTIG, 2026-11-16). GTIG separately assesses that the open-source agentic pentest frameworks it tracks, among them ARTEX, CAIRN, CyberStrikeAI and PentAGI, currently trade stealth for velocity: they generate forensic traffic, rely on recognizable identifiers and leave distinct code-level traces (GTIG, 2026-11-16).

In the past quarter, GTIG identified threat actors using LLM frameworks to fully automate vulnerability discovery and exploitation without a human in the loop for the first time.

It compares code before and after patch announcements, analyzes findings to identify potential vulnerabilities, drafts exploits, and automatically executes them without human intervention.

so that when users started up Claude Code, it would automatically trigger the download of the SOMBERMEME backdoor

Google Threat Intelligence Group / Mandiant 2026-11-16
annual-report17 Nov 04:41Zsingle-sourceOpen finding →

04Updates to prior coverage2 items

HIGHCVE-2026-84869exploitedupdatedNATOA1

CVE-2026-84869, ConnectWise ScreenConnect: a missing file-transfer authorization check lets an active remote session push and auto-run files on the Host, and Huntress traced worm-like exploitation back to 20 August, weeks before any patch existed (CVSS 9.9)

First published 2026-09-12 · open finding →

Updaterun 2026-11-17T0355Z-intelentitiessourcesevidenceactionsbodytitletechniques

Huntress confirmed exploitation of CVE-2026-84869 against MSP technicians: on 2026-11-12 a guest-pushed batch file ran on a technician's client with no prompt and installed a rogue ScreenConnect client within about nine seconds, while attacker guests were making 688,000 run-file requests on 42 MSP servers between 2026-10-20 and 2026-11-16. Updating the server does not fix it; every technician host client must be on 26.6.5 or later.

Huntress confirmed exploitation of CVE-2026-84869 against MSP technicians in a campaign it separates from the crypto-exchange-named decoy sessions it reported on 2026-11-13, and it describes the SCKit malware as appearing to leverage a vulnerable server-side extension, not to exploit this flaw directly (Huntress, 2026-11-16). On 2026-11-12 a technician at one MSP joined a blank-named Access session on the MSP's server and the host client ran a batch file the guest had pushed, with no prompt; within about nine seconds a hidden PowerShell relaunch, a curl download and an administrative-image unpack of an MSI ended with a rebranded ScreenConnect client connected to an attacker relay on port 8041 (Huntress, 2026-11-16). Huntress read the delivery side from server databases: between 2026-10-20 and 2026-11-16 attacker guests made 688,000 run-file requests for update.bat-style files from 29 source addresses on 42 MSP servers, repeating them in case a technician joined, and Huntress assesses with high confidence that the spray operator and the relay operator are the same (Huntress, 2026-11-16). In October an attacker also joined three real customer support sessions as an extra guest and asked the technician's client to run a path-traversal file into the Startup folder; two of the three technicians were on vulnerable Windows clients and the Mac client failed (Huntress, 2026-11-16).

The vulnerable code runs on the technician's machine, so updating the server alone does not close it, and Huntress saw many technicians on 22.x, 24.2 and 25.8 clients weeks after 26.6.5 shipped (Huntress, 2026-11-16). Server-side, Huntress points to guest-originated file-run events (RanFiles, EventType 85, or SentFiles, 86) in App_Data\Session.db, read from a copy, that overlap an open host connection; on workstations it points to the ScreenConnect client running batch or script files from its Temp folder, and to any client whose command line names a relay the organisation does not own (Huntress, 2026-11-16).

HIGHCVE-2026-126147 +18updatedNATOA2

Fortinet coordinated security hardening release: the advisories published on 2026-11-10 carry unauthenticated FortiMail, FortiAnalyzer, FortiManager, FortiOS and FortiAuthenticator flaws rated CVSS up to 9.8; FortiAnalyzer and FortiManager 8.0 have no fixed build for the JSON API flaw, FortiOS 7.0 and FortiMail 7.2 are left out of the October fixes

First published 2026-10-23 · open finding →

Updaterun 2026-11-17T0355Z-intelcvesevidencebody

Fortinet revised the FortiOS HA advisory for CVE-2026-126259 on 2026-11-16: FortiGate HA clusters running FGCP, in active-passive and active-active mode, are also affected on the vulnerable builds, though exploitation needs an attack vector adjacent to the HA segment. The advisory adds an HA hardening configuration for fixed builds whose HA segment is not fully trusted; the fixed builds are unchanged.

Fortinet revised the FortiOS HA advisory on 2026-11-16: FortiGate HA clusters running FGCP, in both active-passive and active-active mode, are also affected on the vulnerable builds, in addition to the FGSP with standalone-config-sync and autoscale deployments named in its summary, but exploitation requires an attack vector adjacent to the HA segment (Fortinet PSIRT, 2026-11-10). For fixed builds where the HA layer-2 segment is not fully trusted, the advisory adds a hardening configuration: an HA password with encryption and authentication enabled for FGCP and standalone-config-sync, an encryption setting and a pre-shared secret on the standalone-cluster object for FGSP peers, and a pre-shared secret on the autoscale object (Fortinet PSIRT, 2026-11-10). The fixed builds are unchanged (8.0.1 or above, 7.6.8, 7.4.13 and 7.2.14, with 7.0 migrating) and the revision says nothing about exploitation.

05Action items5 items

Verification & coverage notes1 run

2026-11-17T0355Z-intel · Sonnet 5.5 · window 26 h · 6 entries published

Verification & coverage notes

Window: gap 24.0 h (daily cadence, previous intel fire 2026-11-16T0356Z), window 26 h. No CISA KEV addition fell inside it (work/2026-11-17T0355Z-intel/kev-window.txt, catalogue 2026.11.13) and no covered CVE newly flags ransomware use. The outage-backfill duty did not apply.

Published: six new entries and two update records. New: OptimJack poisoned vendor scripts (high, University of Basel capture), TGR-STA-0049 PerfectPioneer and KUKURAT (notable), Symantec's AI-assisted retailer intrusion (notable, first coverage from the symantec-security-com lookback), the GTIG AI Threat Tracker (annual-report, notable, only the new findings), BundID mass logins (routine, no Swiss portal reported), and Craft CMS CVE-2026-123488 (notable, first coverage from the cccs-alerts and vulncheck lookbacks; exploitation attributed to CCCS and VulnCheck, not asserted). Updates: the Fortinet hardening entry (FG-IR-26-273 now covers FGCP HA clusters; main analysis cut to stay under 550 words) and the ScreenConnect CVE-2026-84869 entry (Huntress confirms exploitation against MSP technicians; actions re-derived). No deep dive: no candidate cleared the bar.

  • Single-source: the Unit 42, Symantec and GTIG entries rest on the vendor holding the telemetry; the Craft CMS exploitation claim rests on CCCS's unnamed open-source reporting plus VulnCheck's KEV flag and is worded as reported. The BundID entry's credential-stuffing label is Handelsblatt's, not the BSI's.
  • borderline-drop: Dassault 3DEXPERIENCE Station Launcher CVE-2026-93664, CVE-2026-19732 and CVE-2026-77033 (unauthenticated RCE rated 10.0, no fix named, no exploitation), no constituency footprint established; doubt about relevance resolves to drop.
  • borderline-drop: NoisyTCP CVE-2026-99288, denial of service only, about 14 hours on a quiet bare-metal host in the researcher's test, no exploit, kernel fix rides the regular cycle.
  • borderline-drop: SharePoint CVE-2026-124223 and CVE-2026-124222 added to the November table on 2026-11-16, authenticated 8.8, Exploitation Less Likely, not exploited or disclosed. Control Web Panel CVE-2026-124020 and CVE-2026-124021, hosting-panel product, no public-sector footprint, no exploitation. Zyxel CPE CVE-2026-79614, WAN access off by default, ISP-customised models excluded, no exploitation. Splunk SVD-2026-1101, advance notice with no vulnerability content until 2026-11-18.
  • borderline-drop: NCSC.ch post 13180 on the six Danfoss AK-SM 800A flaws and its 13054 edit for XWEB EVO, and its N-central rating change to actively exploited, the Danfoss, XWEB EVO and N-able entries already carry these facts, and "vendor firmware updates are available" would read as a fix where Midnight Blue and the BSI say the latest firmware is still affected.
  • borderline-drop: Microsoft 365 outage of 12 to 14 November and the outage-lure phishing campaign (Microsoft Threat Intelligence, 2026-11-16), availability and dependency event with minor Swiss federal effects; the lure is blocked by Defender for Office 365, has no confirmed credential capture and no actor.
  • borderline-drop: the ARTEX reports from Amazon, Cloudflare and Talos (2026-11-13), saturated story with 13 sources in the store, primaries before the 72 h developing window, marginal delta. Barracuda's ARTEX-pattern cloud intrusion (primaries 2026-11-10 to 11-13, out of window, action only for notified customers). Project Zero's "Full X" policy and the Proofpoint ITM macOS flaw (awareness, no technique; no CVE or fix).
  • borderline-drop: Entre Bièvre et Rhône communal breach, Greiz-Ronneburg hospital extortion, Maileva (22 million recipients), Quantum Systems QBase update compromise, the Andropov hoax e-mails, incident floor or out of nexus: no vector, actor or transferable behavior beyond the impact, no Swiss party, or a mechanism that rests on relayed paywalled reporting.
  • borderline-drop: Aargau CyRA pilot, Vaud CYBER26 exercise, CNIL's territorial-authority sanctions (2026-11-09/10) (awareness or outside the window; Cisco's eMudhra advisory v1.12 and Fortinet's Node2Shell advisory revisions) no impacted product, no material delta.
  • Coverage backlog: two rows added, both held on named conditions with expiry 2026-12-01 (TYPO3-PSA-2026-002 pre-announcement, whose release published at 11:00 UTC after this fire began; Oracle's Critical Security Patch Update due 2026-11-17 after this fire). No earlier row was open.
  • Next fire: Oracle's release (pre-release notice of 2026-11-12: 652 patches, Fusion Middleware 53 unauthenticated rows up to CVSS 10.0, E-Business Suite 22 unauthenticated rows up to 9.8) and TYPO3's advisories land after this fire; the 2026-11-18 fire must read the risk matrices and the TYPO3-CORE-SA pages before dropping any row, and Splunk publishes its advisories on 2026-11-18.
  • Verification: four iterations, low-residual early exit on iteration 4 (truth 0, editorial 1, advisory 1, residual 1). Iteration 1 found 11 truth defects, mostly attribution drift in the new entries (OptimJack's capability and dates spread across all four providers, the Ohey citation date, PerfectPioneer's HTTP/2 fallback and blocklist scope, GTIG's stealth-for-velocity sentence) and three in the ScreenConnect update (a count spliced onto another date, a nine-seconds start, an HVCI inference in a pre-existing sentence); all were fixed. Iterations 2 and 3 narrowed to small residuals (Ohey dating, HTTP/2-only variants, the aggregator's BSI attribution of the credential-stuffing label, router scope, an SCKit clause). Iteration 4's editorial finding (the ScreenConnect entry had no Exposure or Detection line and an unrevisited ATT&CK mapping) was fixed after the pass and not re-verified. Left as advisory by design: Craft's actively-exploited tag and exploited status (attributed to CCCS and VulnCheck in the body, CISA KEV absent), the file names of legitimate signed executables in the Symantec and Unit 42 bodies, and the ScreenConnect action that overlaps one step of the SCKit entry's action. The next quality audit should give the ScreenConnect and Fortinet entries an independent pass.
  • Candidate sources not added: Dassault Systèmes advisories listing and Craft CMS GitHub advisories (single-item yield), enea-insights (telecom-only relevance).
  • Source health: 321 records probed; one flagged (anssi-cyber-gouv-actualites, challenge shell), repaired with a sitemap recipe and clean on the re-probe.
  • Process notes: the reader pool is dead (402 on every key, 401 on the anonymous tier) and extract on a 403-walled host ends in a chain of reader errors before it fails; WebFetch's summariser refused an exploit-kit research page, so such pages read through extract.
  • Coverage gaps: sygnia (403 on every transport), belden-psirt (403 on every transport), lab52 (feed reset, WebFetch 503), connectwise-trust-center (bulletin page 403, GitHub Atom feed reads), handelsblatt.com (extract returned a script shell; a syndicated copy was read), inside-it-ch Plus articles (paywall), the Splunk and Oracle releases (not yet published).
  • Watchlist: products checked=0, hits=0; suppliers checked=0, hits=0
  • Essential-coverage: all 21 essential records attempted.