Daily Threat Briefing — 2026-11-03
57 findings53 articles6 sections
- Critical findings
- 0
- High findings
- 0
- New since 06:03 UTC
- 0
- CISA KEV additions
- 0
- First seen in this period
- 53 Articles377 CVEs12,720 IOCs
First seen in this period: news articles, each counted once · CVEs rated CVSS 7.0+ (NVD, GitHub, WordPress) · new abuse.ch indicators (ThreatFox, URLhaus, Feodo Tracker), dated by abuse.ch.
Priority queue
Act on these first
Critical and high findings from this briefing and its live updates. Most severe first, then newest.
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
- Critical severityNewLive update, section VulnerabilitiesSource: CVE
Executive summary · as of 06:03 UTC
The situation
North Korea's TraderTraitor dominates this window. The FBI and U.S. Secret Service attributed the $6.3 billion Binance theft to the group, which makes it the largest cryptocurrency heist on record. Financially motivated activity against crypto platforms continued elsewhere: more than $46M left Brazil's Mercado Bitcoin, Polymarket pledged $3.9M in restitution, and BTCPay Server warned that attackers are draining hot wallets on outdated instances. Several edge and DevOps vulnerabilities are being actively exploited. Two more Ivanti Connect Secure zero-days have been chained in attacks since at least October 1, over 1,000 GitLab servers remain unpatched against a maximum-severity RCE, and Synology warned of a surge in C3RB3R ransomware attacks against NAS devices. Physical-impact attacks also escalated, with a second wave of remote manipulation of supermarket refrigeration systems across Sweden, Belgium and Luxembourg.
Offensive use of AI is now a recurring operational theme. Chinese operators used the open-source multi-agent ARTEX framework against more than 200 organizations in 14 countries. Mandiant documented LLMJacking (hijacked enterprise LLM access) and agentic offensive frameworks. Chainalysis disclosed a breach attributed to "offensive agentic techniques", and Northflank warned customers to rotate all secrets after their AI API keys were abused. Google and partners dismantled the XCPE botnet of more than 1.3 million Android TV boxes, which Google says the manufacturer deliberately backdoored for proxy rental and ad fraud. The BYOD ransomware gang claimed Michigan State University, alleging PII and passwords for 1.03 million current and former students.
No new CISA KEV additions were recorded in the window. NVD published 369 CVEs (40 reviewed here), heavily weighted toward remotely exploitable command-injection flaws with public exploits in WAVLINK routers, plus multi-tenant escapes in Beta9 and hard-coded JWT authentication bypasses in several admin platforms. 106 WordPress component vulnerabilities were logged, including unauthenticated RCE in s2Member and the weDevs Project Manager plugin. abuse.ch ingested 12,423 new indicators across 264 families. The largest set is GitHub-hosted SalatStealer and fake "zapret" lures, followed by ClearFake, AMOS, ClickFix and Mirai infrastructure. Ransomware leak sites added 127 victim entries, led by bulk postings from N0n, Akira, Qilin and settra.
Live updates
16 updates since 06:03 UTC
Critical sandbox-escape and RCE flaws (CVE-2026-44267, CVE-2026-44266) disclosed in the @enspirit/elo npm package
- Critical severity@enspirit/elo JavaScript emitter sandbox escape to code execution (CVE-2026-44267) (opens the source in a new tab)Source: CVEVulnerabilities
The Elo JavaScript emitter in the npm package @enspirit/elo does not filter `.constructor` member access. This allows a sandbox escape to arbitrary code execution in Node.js or the browser (CVSS 9.8). Applications that evaluate untrusted Elo expressions should update and treat affected hosts as potentially compromised.
Xsolis healthcare AI breach impacts at least nine health systems and 1.4 million patients
- High severityXsolis healthcare AI vendor breach hits nine health systems, 1.4M patients (opens the source in a new tab)Source: RSSData Breaches & Leaks
A breach at Xsolis, a healthcare AI vendor providing utilization management and care coordination technology, has affected at least nine health systems and 1.4 million patients. The company attributed the incident to a targeted phishing attack reported on June 5. Health systems using Xsolis should review their third-party data exposure and notification obligations.
- High severitySouth Korea: personal data of president and trip entourage leaked (opens the source in a new tab)Source: RSSData Breaches & Leaks
South Korea's presidential office convened an emergency meeting over the leak of personal information belonging to President Lee Jae Myung and members of the entourage on his June overseas trip. The source and method of the leak have not been detailed in the report.
- Source: RSSData Breaches & Leaks
CSX, one of the largest U.S. transportation companies, experienced a cyberattack that gave unauthorized parties access to employee information, potentially affecting tens of thousands of people nationwide. Details on the attack vector and attribution have not been disclosed.
- Source: RSSData Breaches & Leaks
A second Liverpool hospital is investigating two potential instances of staff improperly accessing medical records linked to the Southport terror attack. The case highlights continued insider snooping risk around high-profile patient records and the value of access-audit monitoring.
HPE Networking Fabric Composer hit by three critical flaws, including hardcoded CLI credentials enabling unauthenticated admin access
- Critical severityHPE Networking Fabric Composer hardcoded credentials and adjacent RCE (CVE-2026-95731/95732/95733) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-95731 (CVSS 9.8) covers hardcoded CLI credentials in HPE Networking Fabric Composer that let an unauthenticated remote attacker gain administrative access. Two companion flaws, CVE-2026-95732 (OS-level authentication bypass) and CVE-2026-95733 (CVSS 9.6 each), allow unauthenticated adjacent attackers to execute arbitrary code. Prioritize patching and restrict management-plane exposure for these data-center fabric controllers.
- High severityDocker Desktop WSL2 seccomp bypass exposes host via AF_VSOCK (CVE-2026-112917) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-112917 (CVSS 4.0: 9.2) affects Docker Desktop for Windows on x86-64 WSL2. An unprivileged Linux container can use the 32-bit socketcall interface to bypass the default seccomp block on AF_VSOCK, exposing host-facing services and Windows files accessible to the logged-in user. Update Docker Desktop on developer endpoints that run untrusted containers.
- Source: CVEVulnerabilities
fastapi-code-generator versions before 0.8.1 fail to escape OpenAPI paths, operation summaries and response status-code keys when building generated Python code. As a result, a malicious or attacker-influenced OpenAPI spec can inject code into the generated application. The flaws are tracked as CVE-2026-63469, CVE-2026-63470 and CVE-2026-63471 (CVSS 4.0: 9.3); upgrade to 0.8.1 and treat third-party specs as untrusted input.
- Medium severityApple macOS sandbox escapes and kernel memory read (CVE-2026-84633, CVE-2026-112497, CVE-2026-112478) (opens the source in a new tab)Source: CVEVulnerabilities
Apple fixed two app sandbox escapes in macOS: CVE-2026-84633, a path traversal, and CVE-2026-112497, a permissions-checking flaw. Apple also fixed CVE-2026-112478, an out-of-bounds read that can leak kernel memory or crash the system; all three are rated CVSS 9.0. Fixes ship in macOS Golden Gate 27/27.2, Tahoe 26.7.2 and Sequoia 15.8.2.
German lawmakers and foreign-policy staff identified among recipients of Apple's latest mercenary spyware threat notifications
- Medium severityGerman lawmakers and policy staff among targets in Apple spyware warning wave (opens the source in a new tab)Source: RSSThreat Actors
Several German lawmakers and foreign-policy staff were among the recipients of Apple's recent mercenary spyware warnings, which went to targets in 99 countries. This update adds named victimology to the previously reported notification wave and points to targeting of European political and diplomatic personnel. Organizations supporting government or policy staff should check for Apple threat notifications and consider device triage and Lockdown Mode for high-risk users.
Apple patches two CVSS 10 WebContent sandbox escapes across Safari, iOS, macOS and visionOS
- Critical severityApple WebKit Web Content sandbox escape (CVE-2026-113897) (opens the source in a new tab)Source: CVEVulnerabilities
Apple fixed a remotely reachable Web Content sandbox escape (CVSS 10) by removing the vulnerable code. Fixed versions are Safari 27.2, iOS/iPadOS 27.2, macOS Golden Gate 27.2 and visionOS 27.2. Prioritize browser and OS updates, given the mercenary spyware warnings reported earlier today.
- Critical severityApple WebContent sandbox escape across iOS and macOS branches (CVE-2026-112690) (opens the source in a new tab)Source: CVEVulnerabilities
An input-validation flaw (CVSS 10) lets a remote attacker break out of the WebContent sandbox. Fixes span iOS/iPadOS 26.7.2 and 27.2, macOS Golden Gate 27.2, Sequoia 15.8.2, Tahoe 26.7.2 and visionOS 27.2, which shows that older supported branches are also affected.
- Critical severitySolarWinds Dameware command injection (CVE-2026-28336) (opens the source in a new tab)Source: CVEVulnerabilities
An authenticated command injection in SolarWinds Dameware (CVSS 9.9) can lead to compromise of the remotely controlled host. Dameware's privileged remote-administration reach makes this a strong candidate for lateral movement; patch the software and review who has Dameware access.
Microsoft details fake-download campaign using trojanized Inno Setup installers to push monetization payloads
- Medium severityFake-download campaign uses trojanized Inno installers to deliver monetization payloads (opens the source in a new tab)Source: RSSMalware
Microsoft Threat Intelligence is tracking a fake-download campaign that uses trojanized Inno Setup installers to deliver multiple monetization offers (adware/PUP-style payloads). Microsoft provides detection and investigation guidance covering the full installer session, which is useful for hunting installer-initiated child processes and follow-on downloads.
- High severityBinance: $6.3B heist stemmed from months-long campaign using fake job interviews (opens the source in a new tab)Source: RSSThreat Actors
Update to the previously reported FBI attribution: Binance's CEO says the $6.3B theft discovered in October followed a months-long campaign combining multiple techniques, including social engineering through fake recruitment interviews. North Korea has publicly denied responsibility.
Booba Project claims four victims including the City of Edgewater, alleging 47 GB of stolen government data
- High severityBooba Project lists City of Edgewater and three other victims (opens the source in a new tab)Source: BreachData Breaches & Leaks
The Booba Project leak site lists four victims in one batch: the City of Edgewater (government administration, 47 GB claimed), DROUIN SAS (French wood products, 20 GB), Vetstoria (veterinary services, 5 GB) and Boardman Park (recreation, 5 GB). The claims are unverified. The municipal listing is the most consequential because of potential exposure of resident and employee data.
Ziroom ZHOME A0101 router flaw allows remote Lua code injection (CVE-2026-121260)
- High severityZiroom ZHOME A0101 remote Lua code injection (CVE-2026-121260) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-121260 (CVSS 9.1) is a code injection flaw in Ziroom ZHOME A0101 1.0.1.0 smart-home hardware. Unsanitized input reaches loadstring in the Lua handler usr/lib/lua/luci/controller/api/zrUsb.lua, allowing remote code injection. The NVD entry indicates an exploit has been disclosed; restrict management interface exposure and monitor for vendor fixes.
- Medium severityMotionCreative Motion UI WebView arbitrary code execution (CVE-2026-107048) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-107048 (CVSS 9.8) affects the WebView rendering path in the MotionCreative Motion UI Android app (versionCode 101). It allows arbitrary code execution via a crafted POST request. The impact is limited to devices with the app installed.
Barracuda urges patching of two critical flaws that let attackers execute commands as root on its devices
- High severityBarracuda warns of critical flaws allowing root access to devices (opens the source in a new tab)Source: RSSVulnerabilities
Barracuda Networks warned customers to patch two vulnerabilities that can allow attackers to run commands with root privileges on affected devices. The source does not list CVE identifiers or report in-the-wild exploitation. Given past targeting of Barracuda appliances, prioritize patching and review exposed management interfaces.
- Medium severitySentinelOne assesses European refrigeration attacks spanning six countries (opens the source in a new tab)Source: RSSThreat Actors
Update to the previously reported supermarket refrigeration attacks. SentinelOne Labs' assessment says store fridges and freezers warmed overnight across six countries on November 2-3. It names Russian-aligned groups as credible suspects but says attribution remains open.
Critical Kamailio IMS stack overflow (CVE-2026-106803, CVSS 9.8) allows remote exploitation via crafted NOTIFY
- Critical severityKamailio ims_registrar_pcscf stack overflow (CVE-2026-106803) (opens the source in a new tab)Source: CVEVulnerabilities
A stack-based buffer overflow in the ims_registrar_pcscf module of Kamailio v6.1.3 and v6.1.4 lets an attacker send a crafted application/reginfo+xml NOTIFY message. process_contact() then copies an attacker-controlled port value from the alias parameter into a fixed 5-byte stack buffer without validation (CVSS 9.8). Operators of Kamailio-based IMS/VoLTE P-CSCF deployments should prioritize patching and restrict NOTIFY sources.
3AM ransomware affiliate hits UK school estate in suspected AI-assisted intrusion, encrypting ~37 of ~50 reached machines
- High severity3AM ransomware affiliate hits UK school estate in suspected AI-assisted attack (opens the source in a new tab)Source: RSSThreat Actors
A 3AM ransomware affiliate breached a UK school estate and encrypted roughly 37 of nearly 50 machines it reached from an unmanaged host. Researchers found evidence that a model may have generated and revised commands during the intrusion, but stopped short of claiming the attack was fully automated.
- Medium severityRussia-linked cyber spies target Ukraine's war crimes investigations and military aid routes (opens the source in a new tab)Source: RSSThreat Actors
A Russia-linked espionage cluster is using relatively simple malware to steal documents, credentials and other data from Ukrainian organizations. Targets include organizations working on war crimes investigations and military aid routes, particularly those near the front line and the Russian border.
- Medium severityRussia dismantles VasyGrek cybercrime infrastructure, detains three (opens the source in a new tab)Source: RSSPolicy
Russia's Interior Ministry says police have dismantled the infrastructure of the VasyGrek cybercrime group and detained three suspects. The investigation was conducted with assistance from Kaspersky.
- Medium severityEuropean police seize 100+ servers in investment scam crackdown (opens the source in a new tab)Source: RSSPolicy
European law enforcement arrested seven suspects and seized more than 100 servers. The action targeted an unnamed company suspected of supplying fraudulent trading websites to criminal investment-scam gangs.
Revolve says stolen saved-card data may affect up to 1.5 million shoppers, per a Texas breach filing
- Source: RSSData Breaches & Leaks
Fashion retailer Revolve disclosed that shoppers' saved payment card details were stolen. A Texas filing indicates up to 1.5 million people may be affected. Card-fraud teams and issuers should watch for downstream misuse.
- Medium severityHarvard alumni systems breach exposed 851,000 accounts (opens the source in a new tab)Source: RSSData Breaches & Leaks
Have I Been Pwned added data from the November 2025 breach of Harvard University's Alumni Affairs and Development systems, linking it to more than 850,000 unique email addresses. Exposed alumni and donor contacts are likely targets for follow-on phishing and donation-fraud lures.
Orova ransomware group bulk-lists seven victims across Singapore, Malaysia and the U.S., including government-linked MARDEC Berhad
- High severityOrova bulk-posts seven victims, concentrated in Malaysia and Singapore (opens the source in a new tab)Source: BreachData Breaches & Leaks
The Orova leak site posted seven victims in a short burst. They include MARDEC Berhad (formerly a Malaysian government-owned rubber development corporation), Raffles United Holdings/Kian Ho (Singapore), Detronics (Singapore), Bright Property Service (Malaysia), Orient Corporation (fire suppression) and Hydrogel Vision Corp (Florida); the concentration in Southeast Asia suggests a regionally focused campaign.
- High severityRhysida claims 784 GB from Arkansas title and escrow agency (opens the source in a new tab)Source: BreachData Breaches & Leaks
Rhysida claims roughly 2.5M files (~784 GB) from Pope County Title Company and affiliated entities in Russellville, AR, including Johnson County Title, Carol Collins Law and Principle Escrow. Real-estate closing data likely includes PII and banking details useful for wire fraud, which adds to recent title-sector victim listings.
- Source: BreachData Breaches & Leaks
The direwolf group has listed USClaims (usclaims.com), a U.S. litigation-funding finance company. Claimant financial and legal case data may be at risk.
- Medium severityPixelLeaks lists unnamed IT services / hosted email provider (opens the source in a new tab)Source: BreachData Breaches & Leaks
PixelLeaks has posted an unnamed IT services and hosted email provider with a deadline of 2026-11-05 07:22 UTC. A compromise at a hosted email provider carries downstream risk to customer mailboxes, so customers of small hosted email providers should monitor for anomalous access.
Critical unauthenticated API key exfiltration flaw in Eclipse Mnestix Browser (CVE-2026-89292)
- Critical severityEclipse Mnestix Browser unauthenticated backend API key exfiltration (CVE-2026-89292) (opens the source in a new tab)Source: CVEVulnerabilities
Eclipse Mnestix Browser versions 1.6.0 up to (but excluding) 2.3.0 allow an unauthenticated attacker to exfiltrate the MNESTIX_BACKEND_API_KEY for all configured infrastructures (CVSS 4.0: 9.2). Upgrade to 2.3.0 or later and rotate all backend API keys on exposed instances.
Critical WSO2 Identity Server flaw (CVE-2026-4052) lets self-registered users reach WSO2 API Manager System REST APIs
- Critical severityWSO2 Identity Server as Resident Key Manager grants System API access (CVE-2026-4052) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-4052 (CVSS 9.8) affects WSO2 deployments where Identity Server acts as a Resident Key Manager with database sharing. A self-registered user can obtain a token from the Identity Server token endpoint and use it to access WSO2 API Manager System REST APIs. Review whether your deployment uses this configuration, and restrict or disable self-registration until patched.
- High severityManageEngine Log360 and EventLog Analyzer unauthorized access (CVE-2026-13594) (opens the source in a new tab)Source: CVEVulnerabilities
CVE-2026-13594 (CVSS 9.1) affects ManageEngine Log360 and EventLog Analyzer builds 13062 and below and could allow an adversary to gain unauthorized access to the product. These SIEM and log platforms hold sensitive telemetry and are frequent attack targets, so upgrade and restrict management-interface exposure.
Critical WordPress plugin flaws disclosed: ACF Extended PRO code injection (CVSS 9.8) and BuddyPress XProfile Image Field file deletion
- Critical severityACF Extended PRO WordPress plugin code injection (CVE-2026-120534) (opens the source in a new tab)Source: CVEVulnerabilities
The Advanced Custom Fields: Extended PRO plugin for WordPress through version 0.9.2.6 is vulnerable to limited code injection via the render_field function. The flaw stems from insufficient validation of form configuration parameters before they are merged into field settings. Rated CVSS 9.8; update the plugin and review sites for unexpected form or field configuration changes.
- Critical severityBuddyPress XProfile Image Field arbitrary file deletion (CVE-2026-120518) (opens the source in a new tab)Source: CVEVulnerabilities
The BuddyPress XProfile Image Field plugin for WordPress through version 3.1.0 allows arbitrary file deletion. bpxp_image_field_update_on_user_activation() calls unlink() on a path built from visitor-supplied signup metadata without realpath() resolution. Rated CVSS 9.1; deleting files such as wp-config.php can enable full site takeover, so patch or disable the plugin and restrict open registration.
Updates are added during the day when material new intelligence arrives. They are not part of the executive summary; the next daily briefing recaps them.
Browse the full briefing
57 findings · 6 sections- 15 critical
- 25 high
- 14 medium
- 2 low
- 1 info
- Threat Actors8 findingsThreat Actors & Campaigns1 critical · 5 high · 2 medium
- Vulnerabilities17 findingsVulnerabilities & Exploitation13 critical · 4 high
- Malware9 findingsMalware & Indicator Trends5 high · 3 medium · 1 low
- Tools & Techniques5 findingsTools, Techniques & AI Abuse3 high · 1 medium · 1 info
- Data Breaches & Leaks14 findingsData Breaches & Leaks1 critical · 8 high · 5 medium
- Policy4 findingsPolicy, Regulation & Law Enforcement3 medium · 1 low
Threat Actors & Campaigns
Nation-state financial theft, AI-driven intrusion campaigns, an industry-led botnet takedown and hacktivist attacks on alerting and refrigeration systems.
- Critical severityFBI attributes $6.3B Binance heist to North Korea's TraderTraitor (opens the source in a new tab)
The FBI and U.S. Secret Service attributed last week's theft of more than $6.3 billion from Binance to North Korea's state-sponsored TraderTraitor group. This is the largest cryptocurrency heist on record. The FBI is urging the crypto industry to block transactions linked to the stolen funds. Dark Reading's analysis attributes the systemic loss to a single compromised laptop combined with weak system separation. Crypto-sector defenders should screen against FBI-published wallet indicators and review the isolation of signing infrastructure.
Source: RSS - High severityChinese hackers use ARTEX agentic AI framework against 200+ organizations (opens the source in a new tab)
Chinese threat actors used ARTEX, an open-source multi-agent AI penetration-testing framework, to run automated intrusions against more than 200 organizations across 14 countries. Expect faster scanning-to-exploitation cycles against internet-facing services.
- High severityXCPE botnet of 1.3M Android TV boxes dismantled; manufacturer blamed (opens the source in a new tab)
Google and industry partners dismantled the XCPE backdoor and botnet, which spanned more than 1.3 million Android TV set-top boxes. The infected boxes rented out users' internet connections as residential proxies and committed ad fraud. Mandiant assesses with high confidence that the manufacturer, Videostrong, intentionally developed and operated the backdoor for financial gain. Google has banned the TV box maker. Defenders should hunt for unmanaged Android TV boxes on corporate and guest networks.
- High severitySecond wave of attacks manipulates supermarket refrigeration across Europe (opens the source in a new tab)
A second wave of suspected cyberattacks remotely altered temperature settings on freezers and cooling systems at supermarket chains in Sweden, Belgium and Luxembourg, spoiling stock. It follows a similar attack on French stores last month. Retailers should audit internet exposure of refrigeration and building-management controllers.
- High severityBITTER APT deploys PapRAT in South Asia espionage campaign (opens the source in a new tab)
The India-linked BITTER group is targeting South Asian organizations with PapRAT. The campaign uses three distinct attack chains that abuse legitimate Windows functionality.
- High severityApple warns users in 99 countries of mercenary spyware (opens the source in a new tab)
Apple sent threat notifications to users in 99 countries about targeting by mercenary spyware. Apple did not attribute the activity to specific actors or regions. High-risk users should enable Lockdown Mode.
Source: RSS - Medium severityRosstat confirms cyberattack claimed by Belarusian and pro-Ukrainian hackers (opens the source in a new tab)
An alliance of Belarusian and pro-Ukrainian hacktivists claimed to have taken down Russia's federal statistics agency Rosstat and stolen large volumes of government data. Rosstat confirmed a cyberattack.
Source: RSS - Medium severityUkrainian and Russian emergency alert apps hijacked to spread fake nuclear warnings (opens the source in a new tab)
A popular Ukrainian air-raid alert system was hacked to push fake nuclear-attack warnings. A cyber group then breached a similar Russian alert system in retaliation.
Source: RSS
Vulnerabilities & Exploitation
No new CISA KEV additions were recorded. Active exploitation hit Ivanti Connect Secure, GitLab and BTCPay Server. NVD published 369 CVEs (40 reviewed), dominated by IoT command injection, and 106 WordPress component flaws were logged.
- Critical severityIvanti patches two more Connect Secure zero-days exploited in chained attacks (opens the source in a new tab)
Ivanti released updates for two further Connect Secure gateway flaws that attackers have chained in zero-day attacks since at least October 1. Patch immediately and run Ivanti's integrity checker. Treat exposed appliances as potentially compromised and hunt accordingly.
Source: RSS - Critical severityOver 1,000 GitLab servers exposed to actively exploited RCE (opens the source in a new tab)
More than 1,000 internet-exposed GitLab servers remain unpatched against a maximum-severity flaw that is being actively exploited for remote code execution. Alpine Linux rebuilt its GitLab server and reset all passwords after exploitation attempts on October 20-21. Patch, then review GitLab instances for unauthorized runners, tokens and repository changes.
Source: RSS - Critical severityWAVLINK router command-injection cluster (CVE-2026-120345 and related) (opens the source in a new tab)
Many remotely exploitable OS command-injection and stack-overflow flaws with CVSS 9.9-10 were published for WAVLINK routers and APs, with public exploits. Affected models include WN551X3-A, WN536AX6-A, WN553X1-B, WN701AE, WL-WN530BE1-A, WL-WN579A3-A, WL-WN530HG4B and WN579X3-D. The vulnerable components are the WireGuard/OpenVPN CLI handlers, /bin/ioos, protocol.csp and wapi.cgi. Example CVEs: CVE-2026-120345, CVE-2026-120344, CVE-2026-120280, CVE-2026-120240, CVE-2026-120196. These devices are likely Mirai/botnet targets; remove them from internet exposure.
- Critical severityLinksys E8450 and D-Link DIR-505 stack overflows (CVE-2026-120283, CVE-2026-120189) (opens the source in a new tab)
CVE-2026-120283 is a remotely exploitable stack-based buffer overflow (CVSS 10) in the Linksys E8450 portal.cgi hidden_timezone argument. A similar CVSS 10 overflow exists in the D-Link DIR-505 TFTP upload handler (CVE-2026-120189). The vendor did not respond to early contact about the Linksys flaw.
Source: CVE - Critical severityBeta9 multi-tenant isolation escapes (CVE-2026-120306/120307/120308) (opens the source in a new tab)
Three CVSS 9.9 flaws in the Beta9 serverless platform let a tenant break isolation. CVE-2026-120306 allows build-step escape because Buildah runs with chroot isolation. CVE-2026-120307 allows host path traversal via the sandbox file API. CVE-2026-120308 allows a tenant to mount another workspace's volumes.
Source: CVE - Critical severityPrefab UI generative UI sandbox escape (CVE-2026-120338) (opens the source in a new tab)
In prefab-ui 0.14.0-0.20.2, the Generative UI sandbox meant to isolate untrusted (including LLM-written) Python launches Deno with unrestricted permissions, defeating the isolation (CVSS 10).
Source: CVE - Critical severityHard-coded JWT and auth bypass flaws in AJ-Report, ape-volo-admin and Jpom (opens the source in a new tab)
AJ-Report through 1.7.1 signs share tokens with a hard-coded JWT secret, which enables forged tokens and unauthenticated Groovy RCE (CVE-2026-120209). ape-volo-admin before 3.6.1 has a hard-coded JWT signing key that enables admin token forgery (CVE-2026-120442). Jpom 2.10.0-2.11.12 allows super-admin bypass via condition injection in the JPOM-USER-TOKEN header (CVE-2026-120201). All are rated CVSS 9.8.
- Critical severityControl Web Panel OS command injection (CVE-2026-77876) (opens the source in a new tab)
Control Web Panel versions before 1.16 contain an OS command injection via the file POST parameter in the protected_directory module (CVSS 9.8). CWP has a history of rapid exploitation; upgrade promptly.
- Critical severityApache Fineract SQL injection (CVE-2026-64618) (opens the source in a new tab)
Apache Fineract through 1.15.0 has a SQL injection in the outStandingAmountPercentageCondition parameter of the /search/advance API (CVSS 9.8). Separately, CVE-2026-118636 is an unfiltered Java deserialization flaw in the unsupported Apache MINA 3.0 codec; no fix is planned.
Source: CVE - Critical severitys2Member WordPress plugin unauthenticated RCE (CVE-2026-86596) (opens the source in a new tab)
The s2Member WordPress plugin, all versions up to 260927, allows unauthenticated RCE via the item_name parameter because strip_php_tags_deep() sanitization is insufficient (CVSS 9.8).
Source: CVE - Critical severityWordPress Project Manager plugin RCE (CVE-2026-116512) (opens the source in a new tab)
The weDevs Project Manager WordPress plugin up to 4.1.1 allows remote code execution because the 'id' parameter is concatenated without sanitization (CVSS 9.8).
Source: CVE - Critical severityWordPress SimpleForm arbitrary file upload (CVE-2026-118235) (opens the source in a new tab)
SimpleForm up to 3.10.2 allows arbitrary file upload via handle_upload, enabling web shell deployment. Two other CVSS 9.8 plugin flaws were published in the window. Ultra Addons for Contact Form 7 up to 3.5.54 allows unauthenticated arbitrary file deletion (CVE-2026-102435). PDF Invoice for WooCommerce up to 7.1.1 has a PHP object injection via customer_note (CVE-2026-109220). In total, 106 WordPress component vulnerabilities were logged in the window.
- Critical severityUncanny Toolkit for LearnDash authentication bypass (CVE-2026-119096) (opens the source in a new tab)
The Uncanny Toolkit for LearnDash plugin up to 3.8.2 has a parser-differential flaw in uo_login_action() email handling that enables authentication bypass (CVSS 9.8).
Source: CVE - High severityBTCPay Server hot wallets drained on outdated instances (opens the source in a new tab)
BTCPay Server maintainers urged users to upgrade immediately because attackers are draining hot wallets on instances running older versions.
- High severityAndroid "SMS of Death" PoC released (CVE-2026-58882) (opens the source in a new tab)
A public proof-of-concept exploit is now available for CVE-2026-58882, a recently patched Android flaw dubbed "SMS of Death". A crafted text message can remotely disrupt cellular service or crash the phone. Prioritize the latest Android security patch on managed devices.
Source: RSS - High severityWordPress core upgrader vulnerability (CVE-2026-119977) (opens the source in a new tab)
A high-severity (CVSS 8.1) vulnerability affecting the WordPress core upgrader component (class-core-upgrader.php, 7.1.3 branch) was logged. Track the vendor release and keep auto-updates enabled.
Source: CVE - High severity@sd-jwt/decode accepts tampered disclosures (CVE-2026-62374) (opens the source in a new tab)
The npm package @sd-jwt/decode silently drops tampered disclosures instead of rejecting them, which undermines integrity checks in selective-disclosure JWT flows (CVSS 7.1, GHSA-f9j6-8p6x-r9j6).
Source: CVE
Malware & Indicator Trends
abuse.ch ingested 12,423 new indicators across 264 families (8,914 reviewed). Stealer distribution through GitHub, ClearFake/ClickFix social engineering and IoT botnets dominate. Synology reports a C3RB3R ransomware surge against NAS devices.
- High severitySynology warns of C3RB3R ransomware surge targeting NAS (opens the source in a new tab)
Synology urged customers to update NAS devices immediately due to a surge in C3RB3R ransomware attacks. Restrict NAS management interfaces from the internet and verify offline backups.
- High severitySalatStealer and SmartLoader distributed via GitHub "zapret" and VPN lures
URLhaus logged 3,631 GitHub-hosted malware_download URLs tagged SalatStealer/zapret, the window's largest cluster. Related clusters added further URLs: zapret (247), exe/SalatStealer (233), SalatStealer (206), SmartLoader zips (211) and zapret zips (78). The lures pose as zapret Discord/YouTube unblockers, VPN clients and game launchers. Samples: hxxps://raw[.]githubusercontent[.]com/ChuninEnergy/zapret-github-discord/refs/heads/main/service[.]bat, hxxps://raw[.]githubusercontent[.]com/ConquerorSponsor/v2raytun-client/refs/heads/main/v2raytun_windows[.]zip, hxxps://raw[.]githubusercontent[.]com/Baiomystayloriterroristcell609/artun-vpn-premium-loader/refs/heads/main/hypophysics/vpn-premium-artun-loader-v3[.]2[.]zip.
- High severityClearFake fake-update infrastructure (804 indicators)
ThreatFox added 804 ClearFake indicators (domains, URLs, SHA256 hashes) for payload delivery via compromised websites showing fake browser-update prompts. Samples: ob3x7195[.]bossbeyanadolu[.]store, www[.]mirthquakemusic[.]com, vtninfo[.]co[.]nz.
- High severity
- High severityAMOS macOS stealer (646 indicators)
ThreatFox added 646 AMOS (Atomic macOS Stealer) payload hashes and delivery domains. Samples: 99ec22bed70db1293ac55f87019f49d4ce04e2deacb503f498b072b72979d524, 3f84244dcab460b6fba60510836251c5999611c7f440d6bdad5c63121909a9f9, b6d67f2e99a22a3ce5b536b9718d71fef2d2f8cdf2f427e3bd193b318c814e18.
Source: abuse.ch - Medium severityMirai and Mozi IoT botnet infrastructure (802 indicators)
Feeds added 587 Mirai indicators (download URLs, hashes, C2 ip:port) and 215 Mozi indicators. Given the WAVLINK, Linksys and D-Link exploits published this window, expect continued IoT recruitment. Samples: hxxp://117[.]131[.]92[.]150:49257/bin[.]sh, hxxp://103[.]140[.]87[.]105:34247/i, hxxp://112[.]198[.]238[.]18:51961/Mozi[.]m.
- Medium severityVShell and Cobalt Strike C2 servers
ThreatFox added 167 VShell C2 and payload indicators, clustered on 154[.]91[.]52-53[.]x:9097, and 90 Cobalt Strike C2 endpoints. Samples: 154[.]91[.]52[.]169:9097, 80[.]96[.]58[.]218:443, 43[.]243[.]73[.]187:17748.
Source: abuse.ch - Medium severityInfostealer and RAT activity: StrelaStealer, Vidar, PureRAT, WARDENStealer
Feeds added new indicators for four families: StrelaStealer (118 delivery domains), Vidar (105), PureRAT (102, including C2 on 139[.]64[.]172[.]10:56001-56003) and WARDENStealer (59). Samples: hxxps://p[.]au338[.]org/x/, 139[.]64[.]172[.]10:56003, hxxps://knowes[.]homes/.
- Low severityphp.shin web shell infrastructure (88 indicators)
ThreatFox added 88 php.shin_webshell delivery domains, including AWS EC2 hosts and Cloudflare workers.dev subdomains. Samples: ec2-44-249-43-89[.]us-west-2[.]compute[.]amazonaws[.]com, fivokovi[.]workers[.]dev, x2yx9nzu4w[.]workers[.]dev.
Source: abuse.ch
Tools, Techniques & AI Abuse
AI-centric tradecraft: LLMJacking, abuse of stored AI API keys, agentic offensive frameworks and shadow AI-agent incidents.
- High severityMandiant: rise of LLMJacking against cloud and model resources (opens the source in a new tab)
Mandiant responded to multiple 2026 incidents in which actors hijacked enterprise LLM access using stolen cloud credentials and API keys to monetize model usage (LLMJacking). Monitor AI service billing and usage anomalies, and rotate exposed keys.
- High severityNorthflank breach leads to abuse of customer AI API keys (opens the source in a new tab)
Cloud hosting platform Northflank told customers to rotate all credentials and secrets stored on the platform after a breach led to unauthorized use of customers' AI API keys.
- High severityMandiant: ARTEX and agentic offensive frameworks (opens the source in a new tab)
Mandiant analyzes intrusions in which threat actors weaponized agentic offensive frameworks such as ARTEX to run highly automated attacks. The report includes defensive guidance for detecting machine-speed intrusion activity.
Source: RSS - Medium severityDefense contractor connected personal AI agent to secure Pentagon system (opens the source in a new tab)
WIRED reports that a contractor caused a DoD security incident by connecting a secure system used by multiple intelligence agencies to a personal Claude Code account. The case highlights shadow AI-agent risk in sensitive environments.
Source: RSS - Info severityMicrosoft: what exploited vulnerabilities have in common (opens the source in a new tab)
Microsoft finds that real-world exploitation concentrates in a small, high-impact subset of CVEs. It recommends prioritizing patching by exploitation likelihood, exposure and blast radius.
Source: RSS
Data Breaches & Leaks
Leak sites added 127 victim entries (30 reviewed). MSU's alleged credential exposure is the most significant. Crypto platforms, an analytics firm and consumer brands disclosed breaches.
- Critical severityMichigan State University: BYOD claims 1.03M records including passwords (opens the source in a new tab)
The BYOD ransomware gang claims full PII for 1.03 million current and former students and staff, plus current and former passwords and a full system compromise. The claim aligns with a security incident that prompted campus-wide action at MSU. BYOD also hit the University of New Mexico hospital in October. Credential exposure warrants forced resets and credential-stuffing monitoring.
Source: Breach - High severityMercado Bitcoin confirms cyberattack; $46M+ outflow (opens the source in a new tab)
Mercado Bitcoin, Brazil's largest crypto exchange, confirmed a cyberattack. Blockchain analysts tracked more than $46 million leaving the platform.
Source: RSS - High severityChainalysis: data breach via offensive agentic techniques (opens the source in a new tab)
Blockchain analytics firm Chainalysis disclosed that attackers used "offensive agentic techniques" to access internal systems and tools. Exposed data includes customer names, email addresses, user IDs and organizational roles.
Source: RSS - High severityApplebee's: hard-coded app password exposed customer data (opens the source in a new tab)
A researcher found a password embedded in the Applebee's mobile app that exposed customer data. This is a reminder to scan client binaries for hard-coded secrets.
Source: RSS - High severityHolland & Knight listed by Redact (opens the source in a new tab)
The Redact group listed Holland & Knight, a law firm with about $2B in revenue, on its leak site. Law-firm compromises carry downstream client confidentiality risk.
Source: Breach - High severityIsapre Consalud listed by Qilin (opens the source in a new tab)
Qilin listed Chilean health insurer Isapre Consalud and Malaysia's Astino Berhad.
Source: Breach - High severityKingdom Title and others listed by settra (opens the source in a new tab)
The settra group posted kingdomtitle.net, mackinengineering.com and insurancehouse.com, with detailed descriptions of stolen financial records. The Kingdom Title listing cites multimillion-dollar wire transactions and personal financial documents.
Source: Breach - High severityN0n bulk-posts ten anonymized U.S. and Middle East victims (opens the source in a new tab)
N0n posted ten anonymized victims (Company #113-#122) pending exposure. Most are U.S. firms in travel and loyalty, data analytics, identity resolution, dental insurance, biopharma and gaming, alongside UAE and Middle East retailers.
Source: Breach - High severityAkira lists Princeton Management, U.S.NEXT and Howells Motor Freight (opens the source in a new tab)
Akira listed Princeton Management, threatening to upload 130GB of corporate data. It also listed U.S.NEXT, an IT and web services provider with 800+ clients and therefore supply-chain exposure, and Howells Motor Freight.
Source: Breach - Medium severityPolymarket to repay $3.9M after site compromise (opens the source in a new tab)
In its post-mortem of an October 29 site compromise, Polymarket said it will repay users the $3.9 million they lost and compensate them for downtime.
Source: RSS - Medium severitySuntory Holdings: ~100,000 records leaked via cloud service (opens the source in a new tab)
Suntory Holdings reported a leak of personal data on around 100,000 people, including employees, following unauthorized access to an external cloud service.
Source: RSS - Medium severityHYBE breach exposes employee medical and financial data (opens the source in a new tab)
HYBE, the agency behind BTS, disclosed a cyberattack exposing current and former employee and partner data. Exposed fields include resident registration numbers, credit card numbers and medical records.
Source: RSS - Medium severityUNM hospital attackers apologize, offer patients $2,500 (opens the source in a new tab)
The threat actors behind the University of New Mexico hospital attack apologized and offered affected patients $2,500 each, an unprecedented move. They also released new details about the incident.
Source: RSS - Medium severityAbrigo Sageworks fintech platform taken offline after suspected attack (opens the source in a new tab)
Abrigo restored its Sageworks risk management application, except borrower-facing functions, after "irregular activity" forced it offline.
Source: RSS
Policy, Regulation & Law Enforcement
Regulatory enforcement, government threat warnings and prosecutions.
- Medium severityUK NCSC warns of unprecedented rise in cyberattacks (opens the source in a new tab)
In its annual report, the NCSC's chief warns of an "unprecedented rise" in cyberattacks in the coming months. He says long-tolerated vulnerabilities can now be discovered and exploited at unprecedented scale.
Source: RSS - Medium severityHHS OCR settles Lehigh Valley Health Network ransomware investigation (opens the source in a new tab)
HHS OCR settled its HIPAA investigation of the 2023 AlphV/BlackCat ransomware breach of Lehigh Valley Health Network. The attackers had posted patient images on their leak site.
Source: RSS - Medium severitySyrian Electronic Army member Th3Pr0 sentenced (opens the source in a new tab)
Ahmad Umar Agha (Th3Pr0) was sentenced in the Eastern District of Virginia after more than a decade as a fugitive. He targeted Western media, businesses, NGOs and U.S. and allied government entities on behalf of the Assad regime.
Source: RSS - Low severityHong Kong privacy watchdog probes airline and retailer data leaks (opens the source in a new tab)
Hong Kong's Privacy Commissioner is investigating breach notifications involving Hong Kong Airlines, Cathay Pacific, Malaysia Airlines and retailers Yoho and Broadway.
Source: RSS
Sources & Attribution
These briefings are compiled from publicly available threat-intelligence feeds, which may include CISA KEV, NIST NVD, the GitHub Advisory Database (OSV), abuse.ch, and Wordfence Intelligence. Data-breach and credential-leak items may include data from Have I Been Pwned and ransomware.live.
CVE® is a registered trademark of The MITRE Corporation. CVE Records are © The MITRE Corporation, reproduced under the CVE Program Terms of Use. WordPress vulnerability data is provided by Wordfence Intelligence, © Defiant, Inc. Breach data from Have I Been Pwned is licensed under CC BY 4.0.