CTIPilot
AI-generated · no human review · verify critical claims against the linked source. how it works →
‹Fri · 23 Oct 2026
All daily briefs →
Daily brief · UTC day

Friday, 23 October 2026

8 verified findings from 1 run · 4 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. 01Fortinet ships out-of-cycle critical and high fixes for FortiOS and four more products; CVEs wait until 10 November. Fortinet announced on 2026-10-21 a coordinated release of critical and high-severity fixes for FortiOS, FortiManager and FortiAnalyzer (scheduled for 2026-10-22, after recent FortiMail and FortiAuthenticator releases), found through its AI-assisted testing, and urges every customer to upgrade at once, internet-facing devices first. The CVE advisories follow only on 2026-11-10, so there are no identifiers to triage by; NCSC.ch lists the exploitation status as unknown, and the FortiMail builds are the ones that already fix CVE-2026-104286, which Fortinet reports as exploited. →
  2. 02Arista confirms exploitation of a VeloCloud Orchestrator flaw: any tenant login, even read-only, reaches operator rights. Arista's Security Advisory 0199 of 2026-10-22 fixes CVE-2026-114625 (CVSS 3.1 9.9) in on-premises VeloCloud Orchestrator: any authenticated user, down to the lowest built-in tenant role, can gain operator privileges and reach functions meant for internal use only, and Arista says the flaw is known to be actively exploited. Fixed releases are 5.2.3.18, 6.1.3.10 and 6.4.2.12; hosted and dedicated Orchestrators were already patched, and Arista names no actor, victim count or start date. →
  3. 03FortiSandbox auth bypass (9.6) and FortiSIEM pre-auth RCE (9.4) fixed; FortiSIEM PoC is public, no exploitation listed. Fortinet's PSIRT published two critical unauthenticated advisories on 2026-10-13: an authentication bypass in the FortiSandbox GUI (CVE-2026-107656, CVSS 3.1 9.6; fixed in 5.0.7 and 4.4.10) and a path traversal in the FortiSIEM phQueryMaster service that allows code execution (CVE-2026-106173, 9.4; FortiSIEM 7.5.0 only, fixed in 7.5.1, workaround: restrict tcp port 27918). Fortinet lists neither as known to be exploited, and Horizon3.ai published a write-up and a proof of concept for the FortiSIEM flaw on 2026-10-20. The FortiSandbox ranges include 5.0.6 and 4.4.9, the builds that fixed CVE-2026-25089, one of three FortiSandbox flaws reported exploited in June, so a unit patched for that one is still affected. →

01Active threats, incidents & disclosures3 items

NOTABLENATOB2

MythicRatatoskr: ESET documents a Go backdoor that hides its command server inside the Yggdrasil mesh network, and a UAC-0099 to Sandworm access handoff that ended in a Group Policy wiper attack on a Ukrainian grain logistics firm

ESET reports that on 2025-07-16 UAC-0099 gained access to a Ukrainian grain logistics company through a spearphishing attachment: a scheduled task ran a PowerShell payload, the MATCHBOIL downloader deployed the MATCHWOK backdoor, and the next day a previously undocumented Go backdoor, MythicRatatoskr, arrived, which ESET ties to that chain with high confidence because of the timing (ESET Research, 2026-10-22). Further copies reached other machines through an OpenSSH SFTP server, and a SOCKS5 server appeared on some of them (ESET Research, 2026-10-22). About two months later, on 2025-09-17, the SDelete utility was deployed through Active Directory Group Policy, which ESET calls standard Sandworm tradecraft and assesses with high confidence as Sandworm after UAC-0099 handed off the access; ESET could not say which group operated MythicRatatoskr, and it sets the pair beside the earlier Gamaredon and Turla collaboration (ESET Research, 2026-10-22).

The backdoor is built on the user-space yggstack, so it needs no tunnel interface or privileges to join the Yggdrasil mesh, an encrypted IPv6 overlay in the 200::/7 range: it picks three peers at random from an embedded list of public peers and then speaks JSON over HTTPS on port 443 to a command server that has an address only inside the mesh, and some samples wait ten minutes before connecting (ESET Research, 2026-10-22). It runs as a Windows service with an updater-style name, and it exfiltrates files and screenshots over the mesh without staging them (ESET Research, 2026-10-22). ESET notes that hosting the command server in the mesh may prevent network security tools from detecting the outbound connections to it and that blocking the public peer addresses is a viable mitigation (ESET Research, 2026-10-22).

Triage: ESET notes that legitimate users adopt Yggdrasil for resilient connectivity, research and hobby networking, so a single mesh connection needs a business-owner check, while a mesh peer connection together with a service named like an updater is the stronger combination (ESET Research, 2026-10-22).

The data-wiping attack was executed by deploying the SDelete utility via Active Directory Group Policy

blocking the public peer addresses is a viable mitigation

ESET Research (WeLiveSecurity) 2026-10-22

Builds on: Talos: malware tells the analysing LLM to skip it; the text is plain and therefore detectable

threat23 Oct 03:40Zsingle-sourceOpen finding →
NOTABLECVE-2022-21894 +1exploitedNATOB2

BlackX: Kaspersky finds a BlackLotus-style UEFI bootkit on fully patched Windows machines of Central Asian diplomatic bodies, because the Baton Drop fix does not revoke the old boot-manager certificate until an administrator triggers it

Kaspersky GReAT reports that in early 2026 it found BlackX, a UEFI bootkit similar to BlackLotus, used against diplomatic agencies in Tajikistan, Uzbekistan and Turkmenistan after the attackers held administrator privileges: an installer takes an RC4-encrypted package and its key as command-line arguments and drops files to the system and UEFI partitions that exploit Baton Drop (CVE-2022-21894) to enrol attacker-controlled Machine Owner Keys, so arbitrary EFI files load despite Secure Boot (Kaspersky GReAT, 2026-10-22). Kaspersky ties the exploitation to Baton Drop alone; CVE-2023-24932, the flaw Microsoft's revocation guidance is titled for, appears in its post only as the one a UEFI component in ESET's FishMonger report might exploit (Kaspersky GReAT, 2026-10-22). The bootkit injects a kernel driver, signed with a certificate its issuer has since revoked, that disables process, thread and registry callbacks and ETW providers of security products; the user-mode payloads were a RAT that uses Google Drive, OneDrive or Yandex.Disk for command and control and a sideloaded loader (Kaspersky GReAT, 2026-10-22).

The targeted machines were fully patched. Microsoft's fix revokes the Windows Production PCA 2011 certificate only when an administrator sets a Secure Boot registry flag and runs the Secure-Boot-Update scheduled task (Kaspersky GReAT, 2026-10-22); Microsoft's own article says the mitigations in the updates of 2024-07-09 and later are not enabled by default and the Enforcement Phase date is still to be announced (Microsoft Support, 2026-09-30), and Kaspersky says that phase has not been implemented and the certificate's expiry on 2026-10-19 does not stop the vulnerable binaries loading unless it is on the forbidden list (Kaspersky GReAT, 2026-10-22). Kaspersky attributes the campaign to APT41 with medium confidence, resting on the identifier "xdd" shared with a kernel implant in ESET's June 2026 FishMonger report, DLL sideloading and OLLVM (Kaspersky GReAT, 2026-10-22).

the targeted machines were fully patched

However, this revocation does not happen automatically upon patch installation

we have attributed this attack to the APT41 threat actor with medium confidence

Kaspersky GReAT (Securelist) 2026-10-22

However, these mitigations are not enabled by default.

Microsoft Support (KB5025885) 2026-09-30

Builds on: FishMonger (I-SOON) ports its SprySOCKS backdoor to Windows with a kernel-driver rootkit · ESET details 11 Microsoft-signed pre-0.9 UEFI shims that bypass Secure Boot on any machine…

threat23 Oct 03:39Zsingle-sourceOpen finding →
NOTABLENATOB2

M365 token spray and replay: Huntress tracks a 10-week campaign that replays stolen post-MFA tokens from rented hosting within minutes, registers rogue Entra devices, wipes and re-registers MFA methods and hides mail with inbox rules of one to three characters or punctuation only

Between 2026-07-14 and 2026-09-22 Huntress detected 497 Microsoft 365 sign-in events across 79 victim organizations that replayed stolen sign-in tokens instead of guessing passwords (Huntress, 2026-10-22). The tokens came from adversary-in-the-middle relays behind fake sign-in pages, from device-code phishing and from social engineering during a genuine sign-in; a relay passes the real login page and MFA prompt through and keeps the token Microsoft issues afterwards, so its holder does not face MFA again (Huntress, 2026-10-22). Replay followed within minutes, after 12 minutes on 2026-07-21 and four minutes on 2026-09-22, from one of six addresses in a single rented range used in turn, so blocking one address stops at most one phase (Huntress, 2026-10-22).

A scripted replayer with the user agent node requests tokens with the Microsoft Office and Microsoft Authentication Broker client applications against Microsoft Graph, Exchange Online, SharePoint Online and the Device Registration Service and joins 48 rogue Entra devices named DESKTOP- plus exactly eight uppercase hexadecimal characters, where the Windows default is seven alphanumeric characters; a browser-session replayer uses current Chrome or Edge through the OfficeHome client (Huntress, 2026-10-22). What outlives a password reset is the persistence: in 19 organizations the victim's MFA methods were wiped and a new authenticator registered within three hours, the rogue devices stay in the directory until deleted, and hidden inbox rules with names of one to three characters or punctuation only, created from VPN or other rented addresses, moved mail to RSS Feeds, RSS Subscriptions or Conversation History or deleted it, one keyed on payroll and bank terms (Huntress, 2026-10-22). Conditional Access blocks reported as AADSTS53003 fire only after authentication has succeeded, so they signal a valid credential, and ranking tenants by failed attempts points responders at the wrong ones (Huntress, 2026-10-22).

Triage: a node user agent with an Office client identifier can be legitimate automation, so Huntress focuses on sign-ins that touch user mailboxes or the Device Registration Service from hosting infrastructure, and it says the device-name pattern is rare but not unique to this campaign, so a device should be confirmed before it is deleted (Huntress, 2026-10-22).

Between July 14, 2026 and September 22, 2026, Huntress detected 497 Microsoft 365 sign-in events across 79 victim organizations related to a token replay campaign

In 19 organizations, the victim's existing MFA methods were wiped and a new method registered within three hours of the first sign-in from a campaign address.

A password reset alone does not remove attacker-registered devices or MFA methods.

Huntress 2026-10-22

Builds on: ConnectWise SkyKick app credential used in 162+ Microsoft 365 tenants; disabling it does not… · Calendar-invite phishing spreads: Outlook and Gmail add the lure on their own and keep it after… · Microsoft: Storm-0940 runs the ARTEX agent framework at machine speed; Expel adds an Okta…

threat23 Oct 03:39Zsingle-sourceOpen finding →
Sources: Huntress
HIGHNATOA2

Fortinet coordinated security hardening release: critical and high-severity fixes in FortiOS 8.0.2, 7.6.8, 7.4.13 and 7.2.14, FortiManager, FortiAnalyzer, FortiMail and FortiAuthenticator, with the advisories withheld until 2026-11-10

Fortinet announced on 2026-10-21 a coordinated set of security updates for FortiOS, FortiManager and FortiAnalyzer, scheduled for 2026-10-22 and following recent releases for FortiMail and FortiAuthenticator (Fortinet, 2026-10-21). Fortinet says the releases fix vulnerabilities found through its frontier-AI-assisted testing programme, that they include critical and high-severity fixes, and that customers should treat them as a priority change rather than routine maintenance, because AI tools can help adversaries reverse-engineer patches and build working exploits in hours or days, patch differences show where the weakness sits, and AI can chain lower-severity flaws (Fortinet, 2026-10-21). The public advisories will appear only at the next Patch Tuesday, 2026-11-10 (Fortinet, 2026-10-21). NCSC.ch's advisory describes the fixes as ones that might allow attackers to gain unauthorized control over affected security infrastructure and lists the current exploitation status as unknown (NCSC.ch, 2026-10-22).

Fortinet's tables name these affected and fixed versions: FortiOS 8.0.0 to 8.0.1 (fixed 8.0.2), 7.6.0 to 7.6.7 (7.6.8), 7.4.0 to 7.4.12 (7.4.13) and 7.2.0 to 7.2.13 (7.2.14); FortiManager and FortiAnalyzer 8.0.0 to 8.0.1 (8.0.2), 7.6.0 to 7.6.7 (7.6.8), 7.4.0 to 7.4.11 (7.4.12) and 7.2.0 to 7.2.12 (7.2.13); FortiMail 8.0.0 to 8.0.1 (8.0.2), 7.6.0 to 7.6.6 (7.6.7), 7.4.0 to 7.4.8 (7.4.9) and 7.2.0 to 7.2.9 (upgrade to the 7.4 branch or later); FortiAuthenticator 8.0.0 to 8.0.3 (8.0.4) and 6.6.0 to 6.6.10 (6.6.11) (Fortinet, 2026-10-21). FortiSASE updates automatically, FortiManager Cloud may need a manual update, and public-cloud instances follow their cloud partners' release processes (Fortinet, 2026-10-21). heise online lists the same fixed builds and the same 2026-11-10 disclosure date (heise online, 2026-10-22).

The FortiMail rows are identical to the fix table of Fortinet's advisory FG-IR-26-175, an unauthenticated path traversal in the FortiMail IBE webmail service (CVE-2026-104286) that Fortinet reports as exploited in the wild (Fortinet PSIRT, 2026-10-01). Those FortiMail builds therefore already fix a flaw under attack, and the post does not say whether they carry further fixes; for FortiOS, FortiManager, FortiAnalyzer and FortiAuthenticator no source states exploitation.

We strongly urge every customer to plan and promptly complete these upgrades as soon as the releases are available, prioritizing internet-facing devices.

These releases include critical and high-severity fixes.

public advisories will be made available at the next Patch Tuesday on November 10

Fortinet (CISO Collective blog) 2026-10-21

Builds on: Unit 42: a memory-resident FortiGate implant rides an exploited FortiOS flaw that has no fix on… · Fortinet now names fixed FortiMail builds for an unauthenticated file-write flaw it confirmed…

vulnerability23 Oct 03:38Zsingle-sourceOpen finding →
HIGHCVE-2026-114625exploitedNATOA2

CVE-2026-114625, Arista VeloCloud Orchestrator On-Prem: an SQL injection lets any authenticated user, down to the Enterprise Read Only role, act as a superuser operator, and Arista says it is actively exploited (CVSS 3.1 9.9)

Arista's Security Advisory 0199 of 2026-10-22 covers CVE-2026-114625 (CVSS 3.1 9.9, CVSS 4.0 9.4, CWE-89 SQL injection) in the on-premises VeloCloud Orchestrator (VCO): on an affected system any authenticated user can gain operator privileges and masquerade as a local "superuser operator", which includes access to functionality intended for internal use only and not for remote access (Arista Networks, 2026-10-22). Arista states that it detected the issue and that it is known to be actively exploited, and the advisory names no actor, victim count or start date (Arista Networks, 2026-10-22). The prerequisite is network access to the VCO web interface and a valid session on the VCO portal; the lowest role that suffices is Enterprise Read Only, the lowest built-in tenant role, the interface is exposed by default and no configuration prevents the exposure (Arista Networks, 2026-10-22).

Affected are VCO 5.2.x before 5.2.3.18, 6.1.x before 6.1.3.10 and 6.4.x before 6.4.2.12, and end-of-support releases were not assessed; hosted and dedicated VCO were affected and are already patched, while VeloCloud Gateway and VeloCloud Edge are not affected (Arista Networks, 2026-10-22). Arista's recommended resolution is the upgrade, and for an unsupported release train it points customers to its support centre to discuss upgrade options (Arista Networks, 2026-10-22); its CSAF file records the CVSS 3.1 score and the same interim measures, and describes the flaw as "VeloCloud Orchestrator Remote Code Execution (VCO RCE)" where the advisory page names an SQL injection that grants operator privileges; Arista does not say how the two descriptions relate (Arista Networks CSAF, 2026-10-22).

Triage: a session authenticated as a low-privilege tenant account, Enterprise Read Only above all, that reaches operator-only or internal-only functions is the signal, because legitimate operators act from their own accounts; Arista lists attacker source addresses in its advisory (Arista Networks, 2026-10-22).

This issue was detected by Arista and is known to be actively exploited.

On an affected system, any authenticated user can gain operator privileges and masquerade as a local “superuser operator” user.

A successful attack requires network access to the VCO web interface and a valid authenticated session on the VCO portal.

Arista Networks (Security Advisory 0199) 2026-10-22

Builds on: Arista confirms exploitation of an SD-WAN orchestrator flaw that has no patch for two of its… · Arista patches an actively exploited unauthenticated command-injection flaw in on-prem…

vulnerability23 Oct 03:31Zsingle-sourceOpen finding →
NOTABLECVE-2026-107656 +1updatedNATOA2

Fortinet FortiSandbox GUI authentication bypass (CVE-2026-107656, CVSS 9.6) and FortiSIEM 7.5.0 unauthenticated path traversal to code execution (CVE-2026-106173, CVSS 9.4): both critical, fixed 2026-10-13, neither reported exploited

Fortinet's PSIRT published two critical advisories for security-operations products on 2026-10-13. FG-IR-26-185 describes an authentication bypass through an alternate path or channel in the FortiSandbox GUI that a remote unauthenticated attacker can trigger with crafted HTTP requests, listed as CVE-2026-107656 with CVSS 3.1 9.6 and an impact of unauthorized code or command execution (Fortinet PSIRT FG-IR-26-185, 2026-10-13). FG-IR-26-180 describes a path traversal in the FortiSIEM phQueryMaster service ("arbitrary file write in grpc phQueryMaster") that lets an unauthenticated attacker execute unauthorized code or commands through specially crafted packets, listed as CVE-2026-106173 with CVSS 3.1 9.4 (Fortinet PSIRT FG-IR-26-180, 2026-10-13). Both pages state that neither flaw is known to be exploited, and both carry temporal vector strings with an exploit-code maturity of functional (FortiSandbox) and high (FortiSIEM) that the advisories do not explain (Fortinet PSIRT FG-IR-26-185, 2026-10-13; Fortinet PSIRT FG-IR-26-180, 2026-10-13).

For FortiSandbox, versions 5.0.0 through 5.0.6 and 4.4.0 through 4.4.9 are affected and 5.0.7 and 4.4.10 fix it; 5.2 is not affected; FortiSandbox PaaS 5.0.0 through 5.0.6 is affected, and every PaaS 4.4 version must migrate to a fixed release (Fortinet PSIRT FG-IR-26-185, 2026-10-13). The ranges include 5.0.6 and 4.4.9 (Fortinet PSIRT FG-IR-26-185, 2026-10-13). Defused reported exploitation of three FortiSandbox flaws in June, CVE-2026-39813, CVE-2026-39808 and CVE-2026-25089 (Help Net Security, 2026-06-16), and the build that fixes CVE-2026-25089 is 5.0.6 on the 5.0 branch and 4.4.9 on 4.4 (Fortinet PSIRT FG-IR-26-141, 2026-06-09), so a unit patched for that flaw is not patched for this one. For FortiSIEM only 7.5.0 is affected and 7.5.1 fixes it; 7.6 and FortiSIEM Cloud are not affected, and Fortinet's workaround is to restrict access on tcp port 27918 (Fortinet PSIRT FG-IR-26-180, 2026-10-13).

Horizon3.ai published a deep dive and a proof of concept for the FortiSIEM flaw on 2026-10-20: the phQueryMaster gRPC service listens on tcp 27918 on every interface and does not request a client certificate, so any remote client can reach its SendQueryResult method, whose query-id metadata is joined into a file path without validation, and a NUL byte in it discards the suffix the service appends, which turns the write into an overwrite of a chosen file (Horizon3.ai, 2026-10-20). Horizon3 uses it to replace a health-check script that a cron job runs every 60 seconds, which gives code execution within a minute (Horizon3.ai, 2026-10-20; The Hacker News, 2026-10-20). Fortinet's advisory says the flaw is not known to be exploited (Fortinet PSIRT FG-IR-26-180, 2026-10-13).

may allow a remote unauthenticated attacker to bypass authentication via crafted HTTP requests

Fortinet PSIRT (FG-IR-26-185) 2026-10-13

may allow an unauthenticated attacker to execute unauthorized code or commands via specially crafted packets

Restricting access on tcp port 27918 prevents the issue.

Fortinet PSIRT (FG-IR-26-180) 2026-10-13

Our proof of concept can be found on our GitHub.

We did not find any application logs that would explicitly reveal that this vulnerability was exploited.

Horizon3.ai (CVE-2026-106173 deep dive) 2026-10-20
Improvementrun 2026-10-24T0255Z-intelbodycvesevidenceheadlinesourcessourcing_notesummarytags

A public proof of concept exists for the FortiSIEM 7.5.0 flaw: Horizon3.ai published a deep dive and the exploit for CVE-2026-106173 on 2026-10-20 (Horizon3.ai, 2026-10-20; The Hacker News, 2026-10-20). The fixed build, 7.5.1, and the port restriction are unchanged, and Fortinet's advisory still says the flaw is not known to be exploited (Fortinet PSIRT FG-IR-26-180, 2026-10-13).

Builds on: Defused reports exploitation of three pre-auth FortiSandbox flaws and CISA lists two in KEV… · Fortinet ships out-of-cycle critical and high fixes for FortiOS and four more products; CVEs…

vulnerability23 Oct 04:25Zsingle-sourceOpen finding →
NOTABLECVE-2026-109618 +1NATOA2

CVE-2026-109618 / CVE-2026-109619, NetScaler ADC and Gateway: the PHP2Shell flaws reach the management GUI on the NSIP, management-enabled SNIPs and the cluster IP of every supported version, with fixes planned for 2026-10-27

NetScaler Product Security says NetScaler is affected by the two PHP memory-safety flaws CVE-2026-109618 and CVE-2026-109619 because the NetScaler Management Portal uses PHP, that Citrix is working to upgrade the PHP package, and that all supported software versions are affected (NetScaler Product Security, 2026-10-22). Citrix describes both as critical flaws in PHP that could let an unauthenticated attacker execute code in the context of the affected PHP process, each rated 9.2 under CVSS 4.0: CVE-2026-109618 is memory corruption when PHP handles an out-of-memory condition during request processing, and CVE-2026-109619 is stale request state that persists after memory is released and is reused by a later request (NetScaler Product Security, 2026-10-22). The affected surfaces are the NSIP GUI, any SNIP configured to permit management access and the cluster IP GUI in a cluster; NetScaler Console (service and on premises) and NetScaler SVM are not affected (NetScaler Product Security, 2026-10-22).

Citrix plans to release fixes for all supported versions on 2026-10-27 and, until then, tells customers to restrict GUI and API access on those addresses to trusted administrative hosts or narrowly scoped management networks, to keep management interfaces off the internet behind jump hosts or tight ACLs, and to review every address that permits GUI or API management and verify that it is not broadly reachable (NetScaler Product Security, 2026-10-22). Its WAF signature v186 covers backend applications that run PHP, while signatures that protect the NetScaler management interfaces themselves were still in development (NetScaler Product Security, 2026-10-22).

All supported NetScaler software versions are impacted.

Citrix plans to release fixes to address these two PHP memory safety vulnerabilities for all supported NetScaler software versions on October 27, 2026.

Customers should immediately restrict access to NetScaler management interfaces to trusted, limited networks.

NetScaler Product Security (Citrix Tech Zone) 2026-10-22

Builds on: PHP2Shell: fixed PHP builds are out and attack traffic is already hitting unpatched servers… · Palo Alto: PHP2Shell affects PAN-OS and Panorama; PAN-OS hotfixes due 26 to 30 October · Two exploited NetScaler zero-days: one in default configuration, one on by default on VPN… · Citrix confirms attacks on a new NetScaler SAML flaw that the September fixed builds do not… · Citrix: a new NetScaler SAML overflow, rated Critical, reaches the builds that fixed the…

vulnerability23 Oct 03:38Zsingle-sourceOpen finding →
NOTABLECVE-2026-79844 +18NATOA2

HPESBNW05162, HPE Networking (Aruba) AOS-8 and AOS-10 wireless: 72 CVEs fixed in one bulletin, 19 of them unauthenticated and network-reachable at CVSS 9.0 or above, across controllers, gateways and access points

HPE Networking's bulletin HPESBNW05162 (initial release 2026-10-22) fixes 72 CVEs in the wireless operating systems AOS-8 and AOS-10 for Mobility Conductors, Controllers, Gateways managed by HPE Networking Central, AOS-10 access points (AOS-AP) and AOS-8 Instant access points (AOS-IAP) (HPE Networking, 2026-10-22). By the CSAF's CVSS 3.1 scores, 34 of the 72 rate 9.0 or higher (eight at 10.0, ten at 9.8, eleven at 9.6, one at 9.4 and four at 9.3); the 19 rated 10.0, 9.8 or 9.4 need network access and no privileges, and the other 15 are reachable from the adjacent network (HPE Networking CSAF, 2026-10-22). The named mechanisms include authentication bypasses and command injection in the AOS-10 interface and in the PAPI protocol, a buffer overflow and a command injection in the AOS-8 PAPI protocol that arrive as crafted packets on its UDP port, an authentication bypass in the AOS-8 SSH daemon, and SQL injection in the web-based management interface (HPE Networking CSAF, 2026-10-22). HPE says the flaws were generally found by its internal security research, that it is not aware of public discussion or exploit code as of the release, and that customers are strongly urged to patch; for some of the critical rows exploitation also needs preconditions outside the attacker's control, which the bulletin does not itemise (HPE Networking, 2026-10-22)(HPE Networking CSAF, 2026-10-22).

The bulletin text names AOS-10 10.8.x up to 10.8.1.1, 10.7.x up to 10.7.2.6 and 10.4.x up to 10.4.1.13, and AOS-8 8.13.x up to 8.13.3.0 and 8.10.x up to 8.10.0.23, while the machine-readable CSAF lists wider known-affected ranges, AOS-10 10.8.0.0 to 10.8.1.1, 10.6.0.0 to 10.7.2.6 and 10.3.0.0 to 10.4.1.13, and AOS-8 8.12.0.0 to 8.13.3.0 and 8.10.0.0 to 8.10.0.23, which take in the 10.6, 10.3 and 8.12 branches the text omits, so treat the CSAF ranges as the scope; fixed builds are AOS-10.8.1.2, 10.7.2.7 and 10.4.1.14 and AOS-8.13.3.1 and 8.10.0.24 (HPE Networking, 2026-10-22)(HPE Networking CSAF, 2026-10-22). HPE's workaround is to restrict the CLI and web-based management interfaces to a dedicated layer 2 segment or VLAN, controlled by firewall policies at layer 3 and above with accounting controls (HPE Networking, 2026-10-22).

Authentication bypass vulnerabilities in the affected interface of AOS-10 could allow an unauthenticated remote attacker to run arbitrary commands on the underlying host if certain preconditions outside of the attacker's control are met.

HPE Networking (CSAF for HPESBNW05162) 2026-10-22
vulnerability23 Oct 03:36Zsingle-sourceOpen finding →

03Updates to prior coverage4 items

CRITICALCVE-2026-109618 +1updatedNATOA1

CVE-2026-109618 / CVE-2026-109619, PHP (PHP2Shell): two unauthenticated remote-code-execution flaws in the runtime, fixed on 2026-10-21 in 8.5.12, 8.4.27, 8.3.36 and 8.2.35, with attack traffic already observed and no upstream release for 8.1 and older (CVSS 4.0 9.2)

First published 2026-10-21 · open finding →

Updaterun 2026-10-23T0256Z-intelsummarysourcesevidencesourcing_notebody

BACS says it knows of no attack in Switzerland but that PHP applications of the federal administration were affected and that it has no complete overview of the update status; the host Cyon took more than 100,000 sites offline for hours, a UK minister reports 32 government services unavailable with no confirmed compromise, and AWS reports a first exploit attempt on 2026-10-17. Sansec and Red Hat judge exploitation harder than VulnCheck's exploit suggests, and Palo Alto now schedules PAN-OS hotfixes for 2026-10-26 to 2026-10-30.

BACS told Keystone-SDA and SRF on 2026-10-22 that no attack in Switzerland attributable to the two flaws is known to it, while stressing that unrecognised or unreported incidents cannot be excluded and that it has no reliable figures on affected websites, shops and internal applications (SRF, 2026-10-22; watson.ch, 2026-10-22). It also said that PHP-based applications of the federal administration were affected, that the responsible bodies were told to update, and that it has no complete overview of the update status (watson.ch, 2026-10-22; Inside IT, 2026-10-22).

The shutdowns were large and mostly short. The Basel host Cyon took all of its more than 100,000 customer websites offline for some hours and says everything has run again since the morning of 2026-10-22; the University of Geneva's site was still unreachable that afternoon, and Basel-Stadt still had pages offline at midday and says it found no data outflow (SRF, 2026-10-22). Neuchâtel disabled all PHP-based sites, ne.ch and many commune pages among them, its IT department installed the fix, and the EPFL gave its members access to its services by VPN while its site stayed down (Inside IT, 2026-10-22). Risikomonitor's live view counts 576,549 checked Swiss web servers, 540,169 already patched and 20,231 (3.5 %) still affected, with the PHP version hidden on the other 16,149 (Inside IT, 2026-10-22).

In the United Kingdom, a minister told the House of Lords on 2026-10-22 that 32 government services were unavailable and four partly unavailable, largely from precautionary measures while updates and mitigations were applied, and that the government knows of no confirmed compromise of central government systems, an assessment she extended to the wider public sector (UKAuthority, 2026-10-22; The Register, 2026-10-22). The Register adds that government sites in Germany, France, Belgium, Finland and the Netherlands were also taken offline as a precaution and that the NCSC sent 50,000 warnings (The Register, 2026-10-22).

Attack activity began before the release. Inside IT, citing a post by Cloudflare's chief executive, reports millions of attack attempts, the first at 07:43 UTC on 2026-10-21, more than six hours before the release, and relays that AWS threat intelligence recorded an exploit attempt against a honeypot on 2026-10-17 from a commercial anonymisation network, proofs of concept and attempts by several actors from 2026-10-21, and advises an operator who cannot exclude a compromise to start the investigation at least from 2026-10-17; whether the attempts would have succeeded on real systems is open (Inside IT, 2026-10-22). The Register likewise reports that AWS says attackers were exploiting the flaws four days before disclosure (The Register, 2026-10-22).

NOTABLECVE-2026-91403updatedNATOA1

eMudhra: the private keys of eight intermediate CA certificates were trivially recoverable, so anyone could forge trusted HTTPS certificates; browsers block them, clients with their own trust store or no revocation checking need an update (CVE-2026-111943 in Go)

First published 2026-10-21 · open finding →

Updaterun 2026-10-23T0256Z-intelheadlinesummaryaffected_productscvessourcesevidenceactionsbodyentities

Standalone Java runtimes keep their own trust store: Oracle's Security Alert CVE-2026-91403 (CVSS 3.1 9.1) and the OpenJDK advisory of 2026-10-21 ship the distrust update, the tracker lists the fixed builds, replacing blocked.certs or enabling revocation checking is the stop-gap, and NCSC.ch lists the exploitation status as unknown.

On 2026-10-21 Oracle issued Security Alert CVE-2026-91403 and the OpenJDK project an advisory for the same event, which reaches Java runtimes as well; Oracle says it also applies to sandboxed Java Web Start and applet clients (Oracle, 2026-10-21). Affected are Oracle Java SE 8u511, 8u511-perf, 11.0.33, 17.0.21, 21.0.13, 25.0.5 and 27.0.1, GraalVM for JDK 17.0.21 and 21.0.13 and GraalVM Enterprise Edition 21.3.20, and OpenJDK lists 27.0.1, 25.0.5, 21.0.13, 17.0.21, 11.0.32.1 and 8u504 and earlier (Oracle, 2026-10-21; OpenJDK, 2026-10-21). The tracker lists the fixed Java builds as 27.0.1.0.2, 25.0.5.0.2, 21.0.13.0.2, 17.0.21.0.2, 11.0.33.0.2 and 8u511 (Dennis Jackson, 2026-10-20), so compare the full build string, not the release number, against the affected list; the tracker's 8u511 as fixed sits against Oracle's 8u511 as affected, and the sources do not reconcile the two. NCSC.ch's advisory of 2026-10-22 lists the exploitation status as unknown and the OpenJDK guidance as the temporary mitigation (NCSC.ch, 2026-10-22).

NOTABLECVE-2026-109618 +1updatedNATOA2

CVE-2026-109618 / CVE-2026-109619, Palo Alto Networks PAN-OS and Panorama: the PHP2Shell flaws affect every release, reaching PAN-OS through its management web interface; PAN-OS hotfixes are scheduled for 2026-10-26 to 2026-10-30

First published 2026-10-22 · open finding →

Updaterun 2026-10-23T0256Z-inteltitleheadlinesummaryentitiesaffected_productscvessourcesevidencebody

Palo Alto Networks' revised advisory now also lists the WildFire appliances WF-500 and WF-500-B as affected by the PHP flaws in all versions, schedules hotfix builds of every PAN-OS train between 2026-10-26 and 2026-10-30, and leaves Panorama and WF-500 undated; Cortex XCORE (SaaS) and WildFire Cloud are already remediated.

Palo Alto Networks revised its advisory on 2026-10-23: the WildFire appliances WF-500 and WF-500-B are now listed as affected in all versions, and the footnotes say Cortex XCORE (SaaS) was remediated on 2026-10-21 and WildFire Cloud on 2026-10-22 (Palo Alto Networks, 2026-10-23). The advisory now also gives estimated availability dates for hotfix builds of every supported PAN-OS train, from 2026-10-26 (for example 10.2.18-h13, 11.1.16-h4 and 11.2.10-h16) to 2026-10-30, while Panorama and the WildFire appliances remain with no date (Palo Alto Networks, 2026-10-23). Palo Alto still states that it is not aware of malicious exploitation against its products, and the Threat Prevention coverage and the jump-box restriction are unchanged (Palo Alto Networks, 2026-10-23). The advisory names the management web interface only for PAN-OS and does not say which interface of the WildFire appliances is reachable.

HIGHupdatedNATOA2

PK Softech, software supplier of Swiss pension funds: malware and a data outflow at Publica, with the City of Zurich fund PKZH saying its insured persons' test data was stolen, the Bernese funds BPK and BLVK and Pensionskasse Post also reporting the supplier incident, and the Federal Prosecutor's Office investigating

First published 2026-10-09 · open finding →

Updaterun 2026-10-23T0256Z-intelheadlinesummarysourcesevidencesourcing_noteactionsbody

The supplier's written pledge to delete Publica's 2019 migration data was apparently not kept, Inside IT relays that the Federal Prosecutor's Office confirmed Akira exfiltrated data, though that is second-hand and neither the Confederation nor Publica has itself named the actor, and the Ticino cantonal fund IPCT says all its insured could potentially be involved but has no evidence yet that data was taken.

Tamedia's Tages-Anzeiger reported on 2026-10-21 that PK Softech AG had confirmed in writing to Publica that it deleted the data, and that this apparently did not happen: its summary says the company reported in January that it had deleted the information on former Federal Councillors and agents, "but this was false" (translated from German; Tages-Anzeiger, 2026-10-21); the article is paywalled. Inside IT, working from the Tamedia research, dates the written confirmation to January 2026 and says PK Softech confirmed it held no personal data in plaintext and had deleted the data of the 2019 migration, that Publica confirmed the sequence (after the Xplain case it asked the supplier to delete all legacy data, including that of the 2019 migration, and received written confirmation), that Publica is clarifying why personal data presumably lay on the servers regardless, and that PK Softech did not answer and does not comment on individual customer relationships (Inside IT, 2026-10-22). 20 Minuten reports the same sequence and notes that PK Softech received the data in 2019 in a migration for Publica (20 Minuten, 2026-10-22).

Inside IT also relays, from Tamedia, that the Federal Prosecutor's Office confirmed that data was extracted by the Akira ransomware group, which still threatens to publish or sell it; 20 Minuten writes that according to findings so far Akira stole the data (Inside IT, 2026-10-22; 20 Minuten, 2026-10-22). This is a reported confirmation of the actor by an authority, but it reaches the reader second-hand through the paywalled Tamedia article: the Federal Prosecutor's Office is not quoted directly, and Publica's own page, in its entry of 2026-10-20, still says it is clarifying whether data of Publica left and whose (Publica, 2026-10-20).

The Ticino cantonal pension fund IPCT says through Corriere del Ticino, relayed by Ticinonline on 2026-10-20, that potentially all of its insured could be involved but that there is no evidence yet that their data was actually taken; it has filed a complaint with the Federal Prosecutor's Office and is preparing a letter to the insured warning of telephone and e-mail fraud (Ticinonline, 2026-10-20). TicinoNews adds, from its own information, that a folder labelled IPCT with the data of a few hundred insured persons was on the supplier's server, while IPCT, which has about 27,000 affiliated persons, says it has no certainty yet that data of its insured is involved (TicinoNews, 2026-10-21).

04Action items9 items

Verification & coverage notes1 run

2026-10-23T0256Z-intel · Sonnet 5.5 · window 26 h · 8 entries published

Verification & coverage notes

Window: gap 24.0 h from 2026-10-22T0256Z-intel, window 26 h, developing-story window 72 h. The container clock agreed with the network date (skew 0 s) and the first fetch of origin/main was current. A Friday-morning fire in the second day of the PHP2Shell story: the new signal is downstream (vendor impact statements, the Swiss and UK public-sector picture, diverging exploitability assessments), a vendor-confirmed exploited SD-WAN controller flaw, two out-of-cycle vendor releases (HPE Aruba, Fortinet), and three identity, boot-chain and C2 tradecraft reports.

KEV sweep (work/2026-10-23T0256Z-intel/kev-window.txt, catalogue 2026.10.21): no KEV addition inside the window, nothing NOT COVERED, no RANSOMWARE row; S1 read the live catalogue three times and confirmed it. Neither PHP CVE and not CVE-2026-114625 is KEV-listed.

Backlog (state/coverage_backlog.md): no open row, nothing carried or added; the Beyond Gravity row stays struck because S2 found nothing newer than 2026-10-15.

Published (8 new): cve-2026-114625-arista-velocloud-orchestrator-sqli-exploited (high: vendor-confirmed exploitation, any tenant login, no KEV listing; Arista is first-party and the only source, hence single-source; the two earlier VeloCloud entries are declared in references[]), hpesbnw05162-hpe-aruba-aos-8-aos-10-72-cves-19-unauth (notable: 19 unauthenticated network rows at 9.0 or above in one bulletin, HPE knows of no exploitation or public code; clears PD-11(b) through the number of unauthenticated rows on a management plane and campus wireless estates, a one-clause judgement), netscaler-php2shell-management-gui-fixes-2026-10-27 (notable: own entry because the product, the surfaces and the fix date are NetScaler's; declares the PHP entry and the PAN-OS entry in references[]; lookback item, source dated 2026-10-21 20:23 UTC), fortinet-hardening-release-fortios-fortimanager-fortimail (high: out-of-cycle critical and high fixes the vendor urges at once, advisories withheld to 2026-11-10, exploitation unknown; techniques[] carries T1190 on the vendor's statement that internet-facing devices come first and the interim measure restricts administrative interfaces, no CVE ids exist yet), fortinet-fortisandbox-fortisiem-pre-auth-critical-fixes (notable: first coverage of the Fortinet PSIRT bundle of 2026-10-13, found through the verifier's coverage pass, two unauthenticated critical rows, FortiSandbox GUI authentication bypass at 9.6 and FortiSIEM 7.5.0 path traversal to code execution at 9.4, neither listed as exploited; the FortiSandbox ranges include the builds that fixed the three flaws reported exploited in June, which is the non-obvious defender hook; references the June FortiSandbox entry and the hardening entry), m365-token-spray-replay-rogue-entra-devices-mfa-wipe (notable: Huntress only, behaviour-rich identity tradecraft against the Microsoft 365 estate the constituency runs), blackx-uefi-bootkit-baton-drop-revocation-default-off (notable: a state-actor bootkit whose enabling condition is a default-off Microsoft control; actions[] empty because the revocation is a fleet programme, not a do-now task; references the FishMonger and UEFI-shim entries because CVE-2023-24932 appears in the store) and mythicratatoskr-yggdrasil-mesh-c2-uac-0099-sandworm-wiper (notable: Ukrainian victim of 2025, kept for the mesh-C2 detection hook and the access-handoff-to-wiper pattern; ground (d)). Updated (4, all type: update): the PK Softech entry (the supplier's January deletion pledge apparently not kept, a second-hand report that the Federal Prosecutor's Office confirmed Akira, the Ticino fund IPCT), the PHP2Shell entry (BACS: no known Swiss attack but federal PHP applications affected, the UK Lords figures, AWS's 2026-10-17 first attempt as relayed, Sansec's and Red Hat's cautious exploitability assessments against VulnCheck's exploit, vendor impact statements; priority left at critical because attack traffic is observed, authorities order the update and VulnCheck holds a working exploit, with no compromise confirmed), the eMudhra entry (Java: Oracle alert and OpenJDK advisory for CVE-2026-91403, NCSC.ch 13091) and the PAN-OS entry (the advisory's 2026-10-23 hotfix estimates for PAN-OS 12.2 to 10.2, WildFire WF-500 and WF-500-B and Panorama still undated). PHP2Shell on Cisco Expressway stays a one-clause statement on the PHP entry: it is a different finding from the CVE-2026-76466 entry, so no changelog record was written there.

Borderline drops: borderline-drop: YOOtheme Pro 5 image handler CVE-2026-102785 — unauthenticated CVSS 4.0 10.0 but no exploitation reported and no constituency deployment shown; a vendor-ordered immediate update and a package-diff write-up do not clear PD-11 (written, then cut at verification iteration 1); borderline-drop: Splunk October bundle SVD-2026-1007 to 1013 — scheduled bundle, grouped hardening rows with no stated prerequisites, no exploitation, regular patch cycle; borderline-drop: ISC BIND CVE-2026-94300 and two DoS flaws — CVSS 8.1 with high complexity, needs a captured signed GSS-TSIG message, no exploit known; borderline-drop: Roundcube 1.6.20 and 1.7.5 — about twenty fixes, authenticated paths, no exploitation statement; borderline-drop: Chrome 156 and the Edge Chromium batch — no exploitation reported, routine browser cycle; borderline-drop: MikroTik CVE-2026-89035 — unreviewed CVE record plus a BSI marker, no vendor advisory, so a no-fix claim cannot be cited to the vendor; borderline-drop: Siemens October Patch Day via CISA ICSA-26-295 — no exploitation, no constituency deployment shown; borderline-drop: EFK audit of the federal Microsoft 365 tenant — oversight audit, no near-term defender decision; borderline-drop: Swiss Government Cloud tender — procurement, non-security; borderline-drop: Beyond Gravity — no development since 2026-10-15, the struck row's re-open condition is unmet; borderline-drop: Ebury 1.8.3 on 333 Tor exit relays — no constituency nexus (relay operators and hosters); borderline-drop: CVE-2026-99999 PwnEscape (Linux KVM guest-to-host) — needs an attacker-controlled guest, no exploitation, no workaround, regular kernel cycle; borderline-drop: DeadTag expired-tag domains — vendor-driven research, no government nexus, the exploit-kit leg was dropped by earlier fires; borderline-drop: Rapid7 Q3 2026 quarterly report — statistics without a defender takeaway; borderline-drop: Hunt Intelligence Graylog and X-Road activity in Ukraine — Ukraine-only victims, Log4Shell public-PoC chain; borderline-drop: P3 and Navigate360 anonymous-tip breach — no Swiss customer shown, out of nexus; borderline-drop: Railway environment-variable exposure — no vector or actor yet and no Swiss nexus, preliminary report due 2026-10-23 midday ET, a later fire re-gates it; borderline-drop: NEC sub-contractor credential abuse (Japan) — out of nexus, lookback item; borderline-drop: Capcom, Onamae Desktop Cloud, the AfD mail leak and the US, Australian and Japanese incidents S4 listed — out of nexus, no transferable technique; borderline-drop: Ircantec forum-sale claim and Medienwerkstatt (Akira claim, mis-tagged DE) — claims without victim confirmation, PD-6; worth a re-check next fire. out-of-window: Sophos IR vishing plus SSPR abuse with a Primary Refresh Token session — primary dated 2026-10-19, window_hours=26, not a rotational-lookback item; the audit may want it, as it may Talos "Parasitizing trust" (2026-10-21, trend piece, dropped for recency and novelty).

Single-source items and carve-outs: the Arista, HPE, NetScaler and two Fortinet entries are first-party vendor advisories for their own products (NCSC.ch and heise restate Fortinet and add no independent assessment); Huntress, Kaspersky (the campaign and attribution, with Microsoft's KB independently confirming the default-off revocation) and ESET are single research sources; the Arista exploitation statement has no independent corroboration yet. Contradictions and unconfirmed claims reported, not resolved: Inside IT relays Tamedia that the Federal Prosecutor's Office confirmed Akira exfiltrated the data, Publica's own page of 2026-10-20 still says it is clarifying, and no prosecutor text was found; Sansec and Red Hat judge PHP2Shell exploitation harder than VulnCheck's weaponized exploit shows; Huntress gives 303 successes in 29 organizations and later 232 in 28 for one address, so no count from that passage is used.

Coverage gaps: connectwise-trust-center (403 on every transport including WebFetch and the reader, third consecutive run; ConnectWise coverage rests on search and the CCCS feed). Outside this window and with no store entry, for the audit's re-sweep: SAP Security Patch Day 2026-10-13 and the two non-exploited publicly disclosed October Microsoft zero-days. The Fortinet PSIRT bundle of 2026-10-13 (FortiSandbox, FortiSIEM) was first covered by this run after the verifier flagged it. Tages-Anzeiger has no full-text transport (lede and bullets only), so the Tamedia reporting rests on relays. The jina reader pool had three of four keys at HTTP 402 (operational note only). sources.promotion_due was acted on: expel, php-net-news and ssd-disclosure moved to active.

Candidate sources: one added, arista-security-advisories (the PSIRT behind three exploited VeloCloud Orchestrator advisories this year; extract plus the CSAF route). Suggested and not added under the one-per-run guide: isc-bind-security-matrix, ukauthority, 20min-ch, tio-ch, pwn.ai, CBC News Investigates; novee-security and air-security are noted for demotion at the next audit.

Entity registry: four entities registered (campaign:m365-token-spray-replay-huntress-2026, malware:blackx-uefi-bootkit, actor:apt41, malware:mythicratatoskr; attributed-to, affects relations sourced to the entries that state them) and eight product entities synced. BlackLotus is named only as a comparison and stays unkeyed.

Self-identification: this fire's main agent is Sonnet 5.5 (claude-sonnet-5-5) per the harness model line; the four research sub-agents reported Sonnet 5.5 (claude-sonnet-5-5) in their own return headers.

Verification loop: five cti-verification passes, all NEEDS_FIXES, with truth plus editorial counts of 16, 9, 3, 3 and 1 and every claim of every pass walked against the ledger (303, 282, 222, 284 and 165 claims in scope, all answered). Iteration 1 found the stale PAN-OS state (the advisory gained hotfix estimates three minutes after the entry was written), a VulnCheck sentence the page had since reversed, two stale eMudhra statements, the FortiMail overlap with the exploited CVE-2026-104286 entry, a Frontaliere Ticino aggregator citation, the YOOtheme entry (dropped: no constituency nexus, no exploitation), a PHP2Shell update on the Cisco Expressway entry that did not belong to that finding (reverted, one clause stays on the PHP entry), and the Fortinet PSIRT bundle of 2026-10-13 (new entry). Iterations 2 to 4 removed overstatements the fixes themselves introduced or exposed (a location and a decryption condition no source states, an uncited workaround, an exploited status on CVE-2023-24932, undated present-tense statements, store-internal wording, an eMudhra Update section that repeated the main analysis, the HPE exploitation attribution). Iteration 5 left one low-confidence F3 (from unnamed sources, fixed after the pass and not re-verified) and one advisory (extra ATT&CK ids on the Mythic entry, one added). The loop ended by the early exit of the decision rules (truth plus editorial at most 2, no F1 or F4), so verification_residual_count is 1 and no CLEAN confirmation applies; entries_dropped_by_verification is 1 (YOOtheme). One advisory finding was declined with a rebuttal: the NetScaler entry keeps the name PHP2Shell, which Sansec, VulnCheck, watson.ch and The Register use for these flaws, while its own cited source (Citrix) says two PHP memory safety vulnerabilities.