CTIPilot
AI-generated · no human review · verify critical claims against the linked source. how it works →
‹Sun · 18 Oct 2026
All daily briefs →
Daily brief · UTC day

Sunday, 18 October 2026

2 verified findings from 1 run · 3 updates to prior coverage · the settled record for this UTC day, in the classic brief order.

CriticalPatch every internet-facing NetScaler ADC/Gateway now and check for pre-patch compromiseCVE-2026-88771 +7 · exploited · improved 18 Oct 03:35Z
Criticality
Kind
Topic
Region
TL;DR · the day in one read
  1. 01ConnectWise SkyKick app credential used in 162+ Microsoft 365 tenants; disabling it does not undo planted access. ConnectWise says it identified unauthorized access to application identities of its Cloud Backup (formerly SkyKick) service; Huntress and Todyl report that from 15 October the app's credential was used to obtain app-only Microsoft Graph tokens in customer Microsoft 365 tenants; Huntress alone confirmed unauthorized sign-ins in 162 tenants by 16 October and saw ConnectWise Cloud Management apps assign themselves directory roles in about 50 tenants and, in one, new client secrets added to other applications. ConnectWise disabled seven application identities and suspended the service on 17 October, but says disabling does not reverse changes already made in a tenant, so every tenant that ever consented a ConnectWise or SkyKick app needs the apps disabled and its application credentials, directory roles and consents reviewed. How the credential was obtained, who is behind it and what data was read are not public. →
  2. 02Some Rank Math PRO 3.0.123 downloads carried malware; updating to 3.0.124 does not remove what it left. Rank Math says some copies of its WordPress plugin Rank Math PRO 3.0.123, downloaded or updated between 2026-10-15 20:02 UTC and 2026-10-16 07:12 UTC, contained malware that could compromise a site once the plugin is activated, and it released 3.0.124 as the clean build. The malware can recreate a hidden backdoor on page visits and may leave administrator accounts, scheduled tasks and files outside the plugin, so an affected site needs the plugin folder and persistence removed in a set order, a restore or cleanup, and rotation of every credential the site could reach; how the vendor's distribution was compromised and who is behind it are not disclosed. →

01Active threats, incidents & disclosures2 items

HIGHexploitedupdatedNATOA1

ConnectWise Cloud Backup (formerly SkyKick): the vendor's Microsoft 365 application credential was used to sign in to customer tenants, with directory-role self-assignments in about 50 tenants and added credentials in one

ConnectWise says it identified unauthorized access to application identities within its Cloud Backup (formerly SkyKick) environment; on 2026-10-17 it disabled seven application identities, suspended the service, says the shared Cloud Backup credentials are no longer valid, and warns that disabling does not reverse changes already made in a tenant (ConnectWise, 2026-10-19). Cloud Backup, SaaS Security and Cloud Management remain suspended with no confirmed restoration time, which also takes ConnectWise's SaaS Security monitoring and alerting out of service (ConnectWise, 2026-10-19); ConnectWise reports no successful threat activity in the environments it monitors since it implemented mitigations (ConnectWise, 2026-10-19) and says it cannot independently determine whether a given partner or customer tenant was affected (ConnectWise, 2026-10-19). Huntress and Todyl report that on 2026-10-15 the app's identity signed in to customer Microsoft 365 tenants from one hosting-provider address and received app-only Microsoft Graph tokens; the credential sits in the vendor's own tenant, so no user, MFA prompt or user Conditional Access was involved, and the app's consent covers read and write access to mail, files and Teams chats (Huntress, 2026-10-16; Todyl, 2026-10-15). Huntress confirmed unauthorized sign-ins in 162 tenants by 2026-10-16 and later counted almost 1,000 tenants with sign-in activity from that address, over 300 successful, after password spraying against four law-firm tenants on 13 and 14 October, and a separate failed wave targeted 16 tenants in policy and national-security sectors (Huntress, 2026-10-16). In about 50 of its tenants, ConnectWise Cloud Management apps assigned themselves directory roles, mostly Exchange Administrator; in one the Teams app added itself to Global Administrator and client secrets were added to an MSP management application and to Huntress's own application within about five minutes, the credential additions from private Azure space (Huntress, 2026-10-16). Microsoft Threat Intelligence, as BleepingComputer relays it, called it an active supply-chain attack and assessed likely abuse of existing permissions including Directory.ReadWrite.All (BleepingComputer, 2026-10-17). How the credential was obtained, who is behind it and what was read remain unknown (ConnectWise, 2026-10-19; BleepingComputer, 2026-10-17).

Triage: the vendor's apps sign in from Azure addresses with a .NET SDK user agent, while the credential additions used a Python user agent; Python and curl are common in administration, so Huntress says the combination of actor, operation and user agent is the signal (Huntress, 2026-10-16). Role self-assignments carrying the vendor's own user agent are not proven benign, since Huntress has not confirmed vendor automation (Huntress, 2026-10-16).

We recently identified unauthorized access to application identities within the ConnectWise Cloud Backup environment.

Disabling the affected connections prevents new authentication through those connections; it does not reverse changes previously made inside a tenant.

Whether a product vulnerability contributed has not been established

ConnectWise Trust Center (advisory) 2026-10-19

As of 19:00 UTC on October 16, Huntress has confirmed unauthorized sign-ins in 162 tenants.

The credential type was a client assertion, which means the credential lives in the vendor's own tenant.

These tenants are in the policy, national security, trade, and critical mineral sectors.

Huntress 2026-10-16

Because the application already held broad, previously granted permissions in each tenant, those sign-ins gave the attacker the technical ability to read and write mail and attachments, OneDrive and SharePoint files, and Teams chats.

Revoking user sessions does not end this access.

Todyl 2026-10-15

Previously issued access tokens may remain usable until they expire or are otherwise invalidated.

ConnectWise Trust Center (advisory) 2026-10-19

Disabling the app in a tenant stops new tokens from being issued there, though tokens already issued stay valid for up to 90 minutes.

Todyl 2026-10-15

Older tenants may list them under SkyKick names.

SpecterOps 2026-10-17

Since implementing mitigations, we have not observed successful threat activity in the environments we monitor.

ConnectWise Trust Center (advisory) 2026-10-19
Improvementrun 2026-10-19T0256Z-intelactionssourcesevidencesourcing_notebody

ConnectWise's update of 2026-10-18 01:00 UTC took SaaS Security and Cloud Management offline in addition to Cloud Backup, which had been suspended since 2026-10-17, so ConnectWise SaaS Security monitoring and alerting is out of service as well, and the update of 2026-10-19 01:15 UTC still gives no restoration time (ConnectWise, 2026-10-19). Disabling an app stops new tokens only: Todyl and SpecterOps each state that tokens already issued stay valid for up to 90 minutes, and ConnectWise says that after it revoked the Cloud Backup certificates previously issued tokens may remain usable until they expire (Todyl, 2026-10-15; SpecterOps, 2026-10-17; ConnectWise, 2026-10-19). Todyl says to find every SkyKick and ConnectWise app in a tenant, inherited ones included, and that all SkyKick apps are published from one vendor tenant, and SpecterOps hunts for apps owned by that tenant, noting that older tenants may list them under SkyKick names (Todyl, 2026-10-15; SpecterOps, 2026-10-17).

incident18 Oct 03:30Zmulti-sourceOpen finding →
NOTABLENATOA2

Rank Math SEO PRO 3.0.123: some packages served for about eleven hours carried malware that can leave hidden backdoors, rogue administrators and cron persistence that the update to 3.0.124 does not remove

Rank Math says that some copies of Rank Math PRO 3.0.123, downloaded between 2026-10-15 20:02 UTC and 2026-10-16 07:12 UTC, contained malware that could compromise a website if the plugin was activated, and that it removed the files and released 3.0.124 (Rank Math, 2026-10-16). It says the malware can recreate its hidden backdoor whenever someone visits the website, so removing suspicious files before disabling the mechanism that recreates them may not resolve the infection (Rank Math, 2026-10-16). Wordfence adds that the malware may establish persistent access, contact a remote host, and expose WordPress credentials, database credentials, API keys and other secrets that the web server can reach, and that Rank Math says it holds records of who downloaded an affected copy (Wordfence, 2026-10-17). The signs Rank Math lists are three malicious entries in the WordPress options table, hidden PHP backdoors in the must-use plugins directory and in other plugin and theme folders, unknown administrator accounts, a fake core file under wp-includes, .htaccess rules that let .cache files run as PHP, PHP files in the uploads directory and a web-user cron entry carrying a marker comment (Rank Math, 2026-10-16).

We recently identified a security issue affecting a limited number of users who downloaded Rank Math PRO v3.0.123 between October 15, 2026, at 20:02 UTC and October 16, 2026, at 07:12 UTC.

The affected plugin files contained malware that could compromise your website if the plugin was activated.

The malware can recreate its hidden backdoor whenever someone visits the website.

Rank Math 2026-10-16

Simply updating the plugin is not enough because the malware may have placed additional files, administrator accounts, or scheduled tasks elsewhere on the server.

Wordfence 2026-10-17
incident18 Oct 03:31Zsingle-source · victim disclosureOpen finding →
Sources: Rank Math · Wordfence

02Updates to prior coverage3 items

CRITICALCVE-2026-88771 +7exploitedupdatedNATOA1

CVE-2026-88771 / CVE-2026-88772, Citrix NetScaler ADC and Gateway: unauthenticated pre-auth RCE zero-days exploited before a patch existed (CVSS 4.0 9.5)

First published 2026-09-28 · open finding →

Improvementrun 2026-10-18T0256Z-intelentitiesreferencesactionssourcesevidencebody

Zscaler ThreatLabz's first-party exploitation telemetry (2026-10-16) adds post-exploitation behaviour to the compromise check: a Perl CGI handler enabled in the web server configuration besides PHP, and a variant that makes the appliance download and run a 64-bit FreeBSD binary, seen in a burst from at least 26 source addresses; the download server also appeared in its earlier Magento exploitation data, which ThreatLabz calls a lead and not attribution.

Zscaler ThreatLabz published first-party telemetry of exploitation of CVE-2026-88771 on 2026-10-16 (Zscaler ThreatLabz, 2026-10-16). The chain it captured matches the log-poisoning sequence in the sections above: a payload stored in the User-Agent of GET requests to four default logon and VPN paths, a forged packet-engine failure record in the authentication request that searches the access log for the marker, decodes the payload and runs it when the vendor's crash-log script selects that record (Zscaler ThreatLabz, 2026-10-16). Two post-exploitation behaviours are not described above: the payload writes PHP and Perl CGI web shells into six web-served directories and adds a handler that runs .pl files as CGI, besides switching PHP on in the web server configuration, and a shorter variant makes the appliance download a Rust-built 64-bit FreeBSD binary and run it from /var (Zscaler ThreatLabz, 2026-10-16). ThreatLabz saw at least 26 source addresses send the forged record together with that downloader within about six to seven minutes (Zscaler ThreatLabz, 2026-10-16). It says the download server also appeared in its earlier exploitation data for CVE-2026-75650, the Adobe Commerce and Magento flaw it tracks as StyleSmuggler, and calls the overlap an investigative lead that does not establish attribution, because shared hosting, a malware-distribution service or payload reuse by unrelated operators could explain it (Zscaler ThreatLabz, 2026-10-16).

HIGHCVE-2026-105133 +1exploitedupdatedNATOB1

CVE-2026-105133 / CVE-2026-105134, AhsayCBS backup management console: an unauthenticated authentication-bypass and code-execution chain exploited since 7 October to drop webshells and a cryptominer; Ahsay says its earlier patches were incomplete and ships hotfixes 10.3.4.45 and 9.15.2.70 through its partner portal

First published 2026-10-10 · open finding →

Improvementrun 2026-10-18T0256Z-intelsourcesevidenceactionsbody

Horizon3.ai's source-level analysis of 2026-10-14 gives the root cause of both flaws (a caller-set flag that skips the administrator credential check, and an upload path that escapes the folder denylist), shows the chain runs as root on Linux, and lists server-side hunt artifacts: new .jsp files in served application directories, replication receivers created outside a change window, unexpected system administrator accounts, and requests carrying the bypass field or the receiver upload headers.

Horizon3.ai published a source-level analysis of both flaws on 2026-10-14, working from the Linux installer, and says its proof of concept for the chain, from authentication bypass to a webshell, is on its GitHub (Horizon3.ai, 2026-10-14). For CVE-2026-105133, every action of the JSON API extends a base class whose administrator password check is skipped when the request body carries a FromServlet flag that the caller sets, so the JSON API is reachable without credentials, and the AddSysUser action, which creates a system administrator, never consults the caller (Horizon3.ai, 2026-10-14). For CVE-2026-105134, an authentication-bypassed call creates a replication receiver whose home directory the attacker chooses; the system-folder denylist that screens that directory at creation is not applied to the upload path later built from two attacker-supplied headers, in which only the percent and colon characters are escaped, so a traversal sequence walks the write location into a served web application, and one upload of a .jsp file then runs as LocalSystem on Windows and as root on Linux (Horizon3.ai, 2026-10-14). Horizon3 says Ahsay released its hotfixes on 12 October 2026 (Horizon3.ai, 2026-10-14).

HIGHexploitedupdatedNATOA2

Danfoss AK-SM and AK-SC, Copeland xweb and Dixell xweb refrigeration controllers compromised, with forced defrost reported at hundreds of French supermarkets; CERT-FR and NCSC Switzerland say to remove internet exposure

First published 2026-10-13 · open finding →

Updaterun 2026-10-18T0256Z-intelsummaryactionssourcesevidencesourcing_notebody

CERT-FR's alert has a third version dated 2026-10-16 whose guidance differs for compromised and uncompromised units: the whole configuration, defrost ranges included, is to be checked, and a compromised Danfoss unit is to be restarted or otherwise factory-reset and reconfigured, with a caution to check insurance conditions before cutting the contractor's data uplink. The Dixell France operator guide that the alert links says any XWEB PRO on firmware 1.12.1 or earlier must be updated without delay, that XWEB units ship with a factory Admin account, and that remote access must go through an encrypted VPN; neither document names an entry vector.

CERT-FR's alert now has a third version, dated 2026-10-16 and listed on the page as an update of the recommendations; the page does not show which sentences changed (CERT-FR, 2026-10-16). Its current text separates the two cases. For every unit it asks to remove the internet exposure of the affected units and of any unit on a public IP address, and to check the insurance contract conditions before cutting the data uplink to the refrigeration contractor (CERT-FR, 2026-10-16). For a unit not known to be compromised it then asks to check the whole configuration, notably the defrost ranges; for a compromised Danfoss unit it says to restart the equipment or otherwise restore it with a factory reset, and then to reconfigure it for service (CERT-FR, 2026-10-16). In both cases access returns only through a secure VPN or protected network, with source-address filtering for the contractors an emergency measure and not a lasting one, and the maker's practices apply (CERT-FR, 2026-10-16). The alert still says campaigns are probably ongoing, still names no flaw or entry vector, and gives the risk as not specified by the vendor (CERT-FR, 2026-10-16).

The alert's documentation section links an operator guide from Dixell France (Copeland) (CERT-FR, 2026-10-16). The file at that address was replaced on 2026-10-18; its current text, described in the correction of 2026-10-19T04:06:00Z, is Copeland's bulletin of 14 October 2026 followed by Dixell France's own complement (Dixell France, 2026-10-14). That complement says XWEB PRO and EVO ship with the same factory account, Admin with the password Admin, advises keeping the original Admin password sealed or in a password manager and not using it daily, gives each user a named account with an expiry date for an external contractor's account, and requires remote access through an encrypted VPN, with no port forward or DMZ and no remote-control software such as AnyDesk or TeamViewer on a site PC that stays on (Dixell France, 2026-10-14). CERT-FR does not say how the attackers got in, and SecurityWeek writes that neither Danfoss nor French authorities have tied the attack to the flaws Claroty disclosed (CERT-FR, 2026-10-16; SecurityWeek, 2026-10-15).

03Action items11 items

Verification & coverage notes1 run

2026-10-18T0256Z-intel · Sonnet 5.5 · window 26 h · 2 entries published

Verification & coverage notes

Window: gap 24.0 h from 2026-10-17T0256Z-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. The window is a Saturday into Sunday and thin. Two new entries and three changelog records. New: ConnectWise Cloud Backup (formerly SkyKick), high, a vendor-held Microsoft 365 application credential used against 162 or more customer tenants with directory-role self-assignments in about 50 tenants and added credentials in one (S3 and S4 surfaced it independently, merged on one incident key; Admiralty A1); Rank Math SEO PRO 3.0.123, notable, a vendor-confirmed backdoor in packages served for about eleven hours, kept short on precedent (the store carries the ShapedPlugin and Awesome Motive WordPress supply-chain entries) and single-source-victim because Wordfence only relays the vendor. Records: Danfoss and refrigeration controllers (update: CERT-FR's third version of 2026-10-16 changes the compromised-unit sequence to restart first and adds the whole-configuration check, and the Dixell France operator guide gives the XWEB PRO firmware boundary and the factory Admin account; the Detection line, summary and actions were rewritten where they stood), AhsayCBS (improvement: Horizon3's root cause and server-side hunt artifacts, first covered through the horizon3-ai lookback), Citrix NetScaler CVE-2026-88771 (improvement: Zscaler's first-party telemetry adds a Perl CGI handler and a FreeBSD binary delivery; the Magento infrastructure overlap is Zscaler's own lead, not attribution).

KEV sweep (work/2026-10-18T0256Z-intel/kev-window.txt): no CISA KEV addition inside the window (catalogue 2026.10.16 of 2026-10-16 is the newest S1 read), so no NOT COVERED, MENTION-ONLY, COVERED-STALE or RANSOMWARE row. S1 confirmed that no covered CVE moved from not-exploited to exploited and that the ENISA exploited list is fully covered.

Backlog (state/coverage_backlog.md): the one open row, Beyond Gravity, was re-gated on today's facts by S2: nothing newer than the SRF report of 2026-10-15, the vector, behaviour and data class are still not public, so an incident entry would carry no evidence-bound technique. Held unchanged on its named condition (BACS, the company or reporting names a vector, an actor or the data affected), expiry 2026-10-20; nothing changed state, so no note was appended.

borderline-drop lines:

  • borderline-drop: Keycloak 26.8.1 (S1; 19 CVEs) - advisories read are Low or Moderate, EPSS 0.0044 or below, no exploitation or KEV listing, so the regular patch cycle covers it; the SAML HTTP-Redirect signature parser differential has no CVE and rests on an issue title.
  • borderline-drop: Zimbra 10.1.21 advisory table (S1) - the vendor table and heise are 2026-10-01 and 2026-09-29, outside the window, with no exploitation; the password-recovery, WebDAV and OnlyOffice rows carry no CVE or severity, and the covered Zimbra entry already directs readers to 10.1.21. Reopen on Rapid7's next part or a CVE with severity.
  • borderline-drop: October Windows updates default the NTLMv1-derived single sign-on block to Enforce (S3; Microsoft KB5066470) - a vendor hardening default with an outage consequence, no attacker behaviour to map, outside the window; an operations notice, not a hunt, patch or block decision.
  • borderline-drop: IDC Frontier sixth report of 2026-10-18 (S4) - service-status change only on a routine, out-of-region entry; no new cause, route or actor.
  • borderline-drop: FBI PSA I-101326-PSA on North Korean IT-worker facilitators (S2) - US-written, no Swiss nexus, adds little to the three covered DPRK IT-worker entries.
  • borderline-drop: FBI and CISA PSA on election and campaign officials, heise's CDU Heilbronn mailbox hijack, Europol's digital-forensics report, Truesec commentary, Arkadia RAIDEN (S2, S3) - US-midterm or German party focus, no vector or actor, or commentary without telemetry; incident floor.
  • borderline-drop: German universities (Halle-Wittenberg, HU Berlin, LMU), the ChimeraZ burst of French forum claims, the Atexo/Docaposte supplier breach, Eurofiber France, Inter-Etab, ABC Bank, Eichstätt, Brenner Base Tunnel (Everest claim), the French hacktivist DDoS wave, ETS California, Cyber Daily exclusives, BAE Systems (S4) - claims without victim statements (PD-6), no vector or actor (incident floor), or out of nexus.
  • borderline-drop: Elastic Apex Accelerator, Group-IB KGB RAT, Report URI skimmer write-ups, Noma SneakPeek, Any.Run lure statistics, Unit 42's monthly ranking, Volexity part 3, the MALFEX npm package, iVerify recap, The Register on building automation, CSA Lab notes (S3) - no constituency nexus, no new behaviour, or vendor statistics.
  • borderline-drop: Fortinet FG-IR-26-176 to -185, the October Microsoft, Adobe and Palo Alto remainders, Juniper JSA112186, Chrome 155, phpMyFAQ 4.1.9, HPE ClearPass, Splunk SVD-2026-1006 (S1) - none exploited and none KEV-listed; unchanged calls on the facts earlier fires used. Splunk says its advisories publish 2026-10-21, so splunk-advisories should be rotated into S1 on 2026-10-21 or 2026-10-22.

Single-source and carve-outs: Rank Math is single-source-victim (the vendor discloses its own distribution incident and Wordfence relays it; Admiralty A2); the Horizon3 and Zscaler additions are first-party analyses of their own reverse engineering and telemetry and carry no verification change on their entries. Reduced-confidence inclusions: none. Identifier omission: Wordfence's page lists CVE-2026-111862 for the Rank Math incident, but the NVD API returned no record and the CVE Services API a 404 on 2026-10-18T03:27Z, so no cves[] record was written for it.

Contradictions: Huntress's count of 162 confirmed tenants (as of 2026-10-16 19:00 UTC) is the number cited; The Hacker News reports 173 for Huntress and the entry does not repeat it. Todyl's post is dated 2026-10-15 on its page yet references ConnectWise's advisory of 2026-10-16, so it was updated after first publication; the entry cites the page date and says Huntress and Todyl run living posts.

Source-recipe notes (for the next audit): ConnectWise's Trust Center page and Wordfence's blog were read through the metered reader (extract fell to jina); S2 spent about five reader fallbacks, four of them on guessed feed paths, which the source records now describe. The Dixell France guide is undated; the 2026-10-13 date given is the file's last modification on the vendor's server (HEAD request), disclosed in the Danfoss entry's sourcing note.

Coverage gaps: prodaft (no dated listing on any transport), apa-ots (no listing or feed route), hpe-networking-psirt (single advisory readable, no listing), hu-berlin.de notice (Anubis challenge, read only as quoted by borncity).

Essential-coverage: all 21 essential records were attempted and returned content; ncsc-ch-security-hub was an edit scan with no post edited since 2026-10-16T06:56Z.

Excluded as recent attempts: 90 standard or candidate records were held out of the rotation slices because the previous two fires attempted them (belt and braces over the rotation cursor); S4's slice took the twelve oldest-cursor breach and ransomware sources regardless, because that pool is small and entirely attempted by the previous two fires.

Self-identification: main agent Claude Sonnet 5.5 (claude-sonnet-5-5) from the harness-injected model line; all four research sub-agents reported Sonnet 5.5 from their own prompt line.

Verification: four iterations. Iteration 1 (NEEDS_FIXES, 8 truth and 6 editorial findings, 17 with the advisories) and iteration 2 (NEEDS_FIXES, 5 truth) were fixed in place; iterations 3 and 4 are two consecutive CLEAN verdicts from independent cold passes, the second a full-ledger confirmation (141 of 141 claims). After iteration 4 the main agent applied three trivial wording fixes from its advisories (dropped 'paid' from the Rank Math summary, 'the malicious host Rank Math names', a missing verb in the Danfoss update section), none changing a sourced claim, and declined the other five advisories with the reasons recorded on the iteration. The Trust Center page for ConnectWise returned 403 to every live transport in the later passes, so those claims were checked against the body the gate saved at 2026-10-18T03:36Z.

Source health: tools/source_health.py probed 249 records, 247 relevant, one stale (threatpost, a demoted record) and one unreadable; the one UNSOLVED row, znuny-advisories (a candidate that has never been fetched), is a host outage: the site answers HTTP 503 to WebFetch and resets the connection from the container, so the recipe is unchanged and a note says to re-probe next fire.