1. Executive summary
Microsoft's November 2026 Patch Tuesday fixes 1,094 vulnerabilities, and one of them, CVE-2026-124194 in Outlook, was exploited in the wild before the patch shipped. That flaw is rated 6.5 MEDIUM and is in CISA KEV (added 2026-11-10); opening a crafted message file makes Outlook connect to an attacker-controlled location without the external-content warning and leak the user's NTLMv2 hash, which can be relayed or cracked to impersonate the user. Three further flaws carry 9.8 CRITICAL ratings and none is in CISA KEV: CVE-2026-116928 (out-of-bounds write in Windows Search, reachable through an email attachment that Exchange indexes automatically), CVE-2026-119958 (out-of-bounds write in Windows LDAP, reachable against domain controllers) and CVE-2026-112967 (heap overflow in the Xbox Live Networking Service). For EMEA financial services the exposure is concentrated in Exchange, domain controllers and any estate where Outlook is the default client, because the exploited flaw yields reusable credentials rather than code execution, and the two zero-click flaws need no user action at all.
2. Regulatory framing
| Article | Trigger (the fact in this item) | Practical impact |
|---|---|---|
| DORA Art. 17: ICT-related incident management process | CVE-2026-124194 is exploited in the wild and discloses NTLMv2 hashes that can authenticate as the user; Microsoft has not said when attacks started, who is exploiting it, or what they do with the access. | Firms need a detection and response path for hash disclosure from Outlook, not only a patch ticket: hunt for outbound connections triggered by message files, treat any hit as a possible credential compromise and run it through the incident process. |
| DORA Art. 19: reporting of major ICT-related incidents to competent authorities | The exploited flaw and the two zero-click flaws (CVE-2026-116928, CVE-2026-119958) sit in Exchange, LDAP and domain controllers, components whose compromise can meet the major-incident threshold. | Where exploitation is confirmed in a firm's estate, the incident must be assessed against the major-incident criteria and reported on that basis; the patch itself is not a reportable event. |
| NIS2 Art. 21(2)(d): supply chain security measures | The exploited Outlook flaw is patched through the affected Office products, and Microsoft 365 Apps for Enterprise is named as affected, so exposure depends on the update channel a firm's productivity suite is on. | Firms should confirm that their Microsoft 365 and Office update path, including any third-party managed service, actually delivers the Office-side fix rather than assuming Windows Update covers it. |
3. Technical analysis & attack chain
The exploited flaw is CVE-2026-124194, an Outlook spoofing vulnerability rated 6.5 MEDIUM and recorded in CISA KEV with an add date of 2026-11-10. Microsoft marked it as exploited; it did not say when the attacks started, who is exploiting it, or what they do with the access afterwards. No actor is named in any source, and MITRE ATT&CK has no profile tying a group to this exploitation, so attribution is unconfirmed.
Confirmed chain for CVE-2026-124194:
- The attacker sends a specially crafted message file and convinces the user to open it. Delivery is by email or another channel.
- Opening the file causes Outlook to connect to an attacker-controlled network location without showing the usual external-content warning.
- The connection discloses the user's NTLMv2 hash.
- The attacker uses the hash to authenticate as the user. Rapid7 describes two routes: relaying it live against a service that accepts NTLM without adequate safeguards (message signing for SMB or LDAP, Extended Protection for Authentication for Exchange or Active Directory Certificate Services), or cracking it offline to recover the underlying credential.
The Preview Pane is not an attack vector for this CVE, which limits its reach. The 6.5 score reflects the user-interaction requirement; CrowdStrike and Rapid7 both state the real-world exposure is broader than the score suggests because the flaw bypasses a visible warning and yields reusable authentication material. The advisory credits discovery to a researcher who used Anthropic's Claude, with no version noted. All supported versions of classic Outlook receive patches, and the fix is delivered through the affected Office products, with Microsoft 365 Apps for Enterprise and editions of Microsoft Office named as affected.
The two zero-click flaws need no user interaction:
- CVE-2026-116928, 9.8 CRITICAL, out-of-bounds write in the Microsoft Windows Search Component leading to remote code execution. An unauthenticated attacker sends an email with a specially crafted attachment to Exchange; Exchange processes the attachment automatically for indexing, so the recipient need not open the email or attachment, and code runs with the permissions of the affected processing service. Microsoft has not reported exploitation.
- CVE-2026-119958, 9.8 CRITICAL, out-of-bounds write in Windows' LDAP implementation. An unauthenticated attacker sends specially crafted LDAP requests to a vulnerable system and can execute code remotely without user interaction, potentially compromising domain controllers.
CVE-2026-112967, 9.8 CRITICAL, is a heap overflow in the Windows Xbox Live Networking Service that allows remote code execution without authentication or user interaction.
Dustin Childs of TrendAI's Zero Day Initiative counts 30 CVEs in the release that pose a wormable risk (zero authentication, zero user interaction over a network); the list is in his blog, which is not among the sources here. Patch volume is the context: The Register and Krebs report 1,094 fixes, Tenable 1,093, Rapid7 1,092 own-product vulnerabilities plus 1,102 in Windows, and CrowdStrike 1,252 across vendors; the counts differ by how each counts the release. October had 1,474. CVE-2026-124194 is the only one under active exploitation and none is listed as publicly known. CrowdStrike also flags eight Critical RCE flaws in Office applications reachable through the Preview Pane, including CVE-2026-110043 and CVE-2026-110068 in Outlook, and AD DS flaws CVE-2026-118146 and CVE-2026-118148; those are separate CVEs from the four scored here and carry no verified score in this advisory.
Single-source caveat: the NTLM relay and offline-cracking detail rests on Rapid7's analysis, and the Preview Pane exclusion on CrowdStrike and Rapid7 together. No atomic indicators (hashes, domains, IPs, filenames) for the exploitation are given in any source.
4. Mitigation & containment
P1, within 24 hours:
- Patch Outlook through the Office update channel, not Windows Update alone. CVE-2026-124194 is fixed in the affected Office products, and Microsoft 365 Apps for Enterprise is named as affected. Confirm the Office build actually updated on endpoints.
- Patch Exchange for CVE-2026-116928 and domain controllers for CVE-2026-119958. Childs' priority order is Exchange and domain controllers first to stop zero-click compromise.
- Hunt for the CVE-2026-124194 behaviour: outbound connections from Outlook to unfamiliar external hosts immediately after a message file is opened, and NTLM authentication to SMB, LDAP, Exchange or AD CS from hosts or accounts that should not be authenticating that way.
- Enforce the relay safeguards Rapid7 names: SMB and LDAP message signing, and Extended Protection for Authentication on Exchange and AD CS. These close the live-relay route even where a hash leaks.
- Reset credentials and revoke sessions for any account whose NTLMv2 hash is suspected of disclosure, and review authentication logs for that account against other services.
P2, within 72 hours:
- Ring-fence network-exposed DNS and DHCP services, per Childs, to limit wormable spread.
- Patch systems running the Xbox Live Networking Service for CVE-2026-112967.
- Verify manual administrative fixes on cloud services, which Childs flags because automated patching does not always reach them.
P3, within 7 days:
- Work through Childs' list of 30 wormable-risk CVEs from his blog, which is not reproduced here.
- Review the Office Preview Pane exposure CrowdStrike describes (CVE-2026-110043, CVE-2026-110068 and related Office RCE flaws) and decide whether to disable the Reading Pane in high-risk mailboxes.
No fixed release identifiers, KB numbers or registry keys are given in the sources for any of these CVEs; apply the vendor updates and confirm versions against Microsoft's advisories.
5. Indicators of compromise
No indicators of compromise available in the source material. No hashes, domains, IP addresses, filenames or other atomic values for the exploitation appear in the sources.
Behavioural indicators:
| behaviour | where to observe | confidence |
|---|---|---|
| Outlook opens an outbound connection to an external host immediately after a message file is opened, with no external-content warning shown | Endpoint network telemetry and Outlook process connections | Medium; described by CrowdStrike, Rapid7 and TrendAI's Childs, no atomic values given |
| NTLM authentication from a user or host to SMB, LDAP, Exchange or AD CS in a pattern consistent with relay | Domain controller and service authentication logs | Medium; relay route described by Rapid7 |
| Exchange processes an inbound email attachment for indexing without user action | Exchange transport and search-indexing logs | Medium; Microsoft description via The Register and Malwarebytes |
| Specially crafted LDAP requests arriving at a domain controller from an unauthenticated source | Domain controller LDAP service logs and network capture | Medium; Microsoft description via The Register |
6. Detection
Insufficient indicators to author detection rules. The sources describe behaviours but give no strings, filenames, command lines, registry keys or other artefacts of the exploitation to build a rule on.
CVE assessment
4 referenced CVEs — 1 actively exploited (CISA KEV), 3 critical (CVSS ≥ 9.0)
| CVE | CVSS | Exploited | EPSS | Summary |
|---|---|---|---|---|
| CVE-2026-124194 | 6.5 Medium | ⚠ KEV 2026-11-10 | — | Product ui does not warn user of unsafe actions in Microsoft Office Outlook allows an unauthorized attacker to perform spoofing… |
| CVE-2026-116928 | 9.8 Critical | — | — | Out-of-bounds write in Microsoft Windows Search Component allows an unauthorized attacker to execute code over a network. |
| CVE-2026-119958 | 9.8 Critical | — | — | Out-of-bounds write in Windows LDAP - Lightweight Directory Access Protocol allows an unauthorized attacker to execute code ove… |
| CVE-2026-112967 | 9.8 Critical | — | — | Heap-based buffer overflow in Windows Xbox Live Networking Service allows an unauthorized attacker to execute code over a network. |
7. Sources
- The Register, November Patch Tuesday boasts over 1,000 bugs, one already under attack, https://www.theregister.com/security/2026/11/10/november-patch-tuesday-boasts-over-1000-bugs-one-already-under-attack/5308946, 2026-11-10
- CrowdStrike, November 2026 Patch Tuesday: 1,252 CVEs Include One Zero-Day and 68 Critical Vulnerabilities, https://www.crowdstrike.com/en-us/blog/patch-tuesday-analysis-november-2026/
- Tenable Research, Microsoft's November 2026 Patch Tuesday addresses 1093 CVEs (CVE-2026-124194), https://www.tenable.com/blog/microsoft-november-2026-patch-tuesday-addresses-1093-cves-cve-2026-124194
- Malwarebytes Labs, Microsoft Patch Tuesday fixes Outlook flaw already under attack, https://www.malwarebytes.com/blog/bugs/2026/11/microsoft-patch-tuesday-fixes-outlook-flaw-already-under-attack
- Rapid7, Patch Tuesday - November 2026, https://www.rapid7.com/blog/post/em-patch-tuesday-november-2026
- Krebs on Security, Microsoft Fixes Nearly 1,100 Security Vulnerabilities, https://krebsonsecurity.com/2026/11/microsoft-fixes-nearly-1100-security-vulnerabilities/
8. Adverse Trace position
Adverse Trace assesses CVE-2026-124194 as the priority for EMEA financial services despite its 6.5 MEDIUM score and CISA KEV listing, because it is confirmed exploited in the wild and yields reusable NTLMv2 credentials against Exchange, LDAP and AD CS, the components that underpin authentication across a domain. The three 9.8 CRITICAL flaws (CVE-2026-116928, CVE-2026-119958, CVE-2026-112967) are not in CISA KEV and no source reports exploitation, so they rank as urgent patching rather than active incident, with CVE-2026-116928 and CVE-2026-119958 the ones to close first given their zero-click reach into Exchange and domain controllers. No actor is attributed and the post-exploitation use of the stolen hashes is not stated in public reporting; the relay and cracking routes rest on Rapid7's single analysis and should be verified against a firm's own authentication configuration before enforcement. We will track Microsoft's advisory for exploitation timing and any named actor, watch for public exploit code against the three critical flaws now that the patches are diffable, and update this note if atomic indicators for the Outlook exploitation are published.
Published via PulseTrace — Adverse Trace threat intelligence.