Product & Security Tips — SamHouston (Veracode) asked a question.

Veracode Daily Threat Intel Brief | 2026-11-03

Join the VERACODE Community Group - Product & Security Tips - for the latest Veracode Daily Threat Intel Brief - Stay current, stay secure!

https://community.veracode.com/s/group/0F9Uf0000001c01KAA/product-security-tips

 

CISO Advisory Brief | Tuesday, November 3, 2026

Labels: UNCONFIRMED = not independently confirmed; SIGNAL = early indicator; SINGLE SOURCE = one non-official source; CONFLICTING = sources disagree.

Bottom line: A quieter day for new exploited CVEs: CISA added nothing to the KEV catalog between 31 Oct and 3 Nov (catalog read today). What changed is mostly follow-through. Ivanti customers who patched in October must patch again. The Cursor-marketplace Claude Code backdoor covers two versions, not one. A Synology ransomware wave is hitting unpatched NAS boxes. New malicious packages appeared in npm, PyPI and RubyGems. We also correct yesterday's GitLab fix version.

1. Executive Snapshot (next 24-72 hours, in priority order)

1. Re-patch Ivanti Connect Secure to 25.1.1.3 or 22.8R2.6, then hunt off the appliance. The October fix builds (25.1.1.2, 22.8R2.5) are still exposed to CVE-2026-111357 and CVE-2026-118527, chained with CVE-2026-104935 for unauthenticated root. The federal deadline for CVE-2026-111357 passed on 2 Nov, with forensic triage required. CVE-2026-118527 is due 13 Nov. Changed since 2 Nov: two sources now report that the activity evades Ivanti's Integrity Checker (Volexity via BleepingComputer; Rapid7 via Help Net Security), so a clean result is not an all-clear. Review domain-controller logs for LAPS and BitLocker-key queries and rotate the appliance's LDAP service credentials. (CISA KEV; Ivanti advisory; SecurityWeek; Help Net Security)

2. Clean developer machines of the tampered Claude Code extension: both 2.1.312 and 2.1.313. Cursor's marketplace served them on 29 Oct (about 2:27 am-1:35 pm PT). Changed: Cursor's revised notice adds 2.1.312, which is also a genuine Anthropic version number, so the version alone does not prove infection. Check authorized_keys for the svc-telemetry key and run Cursor's remediation scripts (uninstalling is not enough). Then rotate credentials from a trusted machine. (Cursor notice; Microsoft Threat Intelligence; SecurityWeek)

3. Block today's new malicious packages and confirm dependency-confusion controls. The packages are:

  • npm scopes @goparrot-dashboard and @goparrot-dashboard-api: 25 packages published 1 Nov (MAL-2026-18511).
  • PyPI skills-analyzer 0.1.0-0.1.2 and its dependency is-it-ai, a remote-executable loader (MAL-2026-18427).
  • RubyGems dopc-client 1.0.0 and 10.0.0 (MAL-2026-18518).
  • Make sure internal package scopes resolve only from the internal registry. (GitHub Advisory Database / OpenSSF official records)

4. Update every Synology NAS and take DSM management off the internet. Synology warned on 29 Oct that attackers are exploiting a flaw patched in December 2024 (advisory SA-24:20) to encrypt data on outdated DSM. Victims report C3RB3R ransom notes. Synology has not said which CVE is being exploited. (Synology; BleepingComputer; The Hacker News)

5. Fix GitLab to the official versions and close this week's KEV deadlines.

  • Correction: NVD lists the GitLab CVE-2026-97229 fixes as 19.2.8, 19.3.5 and 19.4.2. We found no source for yesterday's "19.5.1" and withdraw it. Its federal deadline passed on 1 Nov.
  • Brotli CVE-2026-111497 (rebuild Node.js and .NET images) and Adobe Commerce CVE-2026-48356 are due 10 Nov.
  • VMware ESX CVE-2026-41709 is due 11 Nov.
  • (NVD; CISA KEV)

 

The full brief continues in the answers below: High-Visibility Breaches, Threat Landscape, Malicious Libraries, Critical Vulnerabilities, Underground and Leak Signals, Foresight, CISO Control Checklist, and Sources.


  • SamHouston (Veracode)

    2. High-Visibility Breaches

    • Binance (crypto exchange) — Theft on or about 29 Oct. CONFLICTING loss figures: FBI ">$6B"; Binance ~$6.3B; TRM ~$6.36B; Chainalysis ~$6.4B. Frozen ~$2.7-3B by source. Withdrawals resumed in phases from 1 Nov; Binance says ~95% of assets covered. Attacker/method: North Korea "TraderTraitor" (FBI, 31 Oct). Entry via a compromised DevOps engineer's laptop, with further access paths found but not detailed. North Korea denies it (SINGLE SOURCE). Confidence: High: FBI and victim confirmed.
    • GMO Internet (Onamae Desktop Cloud, Japan) — Intrusion detected 20 Oct; malware could have been placed on up to 26,739 customers' servers. New 2 Nov (SINGLE SOURCE, Kyodo): attacker present since 5 Aug. Attacker/method: Unnamed. A North Korea link was reportedly given to the Japanese government: UNCONFIRMED. Ransom demand CONFLICTING: ~¥8B/$50M vs $1M. Confidence: Incident confirmed by victim; attribution unconfirmed.
    • Revolve (US retailer) — SEC 8-K: names, contact details and payment card numbers and expiry dates of customers with saved cards may have been taken "through automated means". Attacker/method: No actor named; automated access. Confidence: Confirmed (8-K). Count CONFLICTING: ≥179,259 (four AG notices) vs 1,515,073 in Texas AG data (not publicly displayed).
    • US DHS — Contact and job data on 41,000+ staff published via DDoSecrets (28 Oct). DHS "aware of a potential breach". Attacker/method: Hacktivists "Department of Peace" claim access via an identity service; Okta says there is no evidence it was compromised. Route UNCONFIRMED. Confidence: Leak confirmed; method unconfirmed.
    • UNM Health / Michigan State Univ. — UNM (26 Oct): ambulances diverted, outpatient surgeries and appointments suspended. MSU listed on the BYOD leak site on 1 Nov, claiming data on 1.3M students (post may say 1.03M). Attacker/method: BYOD ransomware gang. UNM disputes BYOD's "36,000 systems". MSU calls it a "potential credential compromise event". Confidence: UNM disruption confirmed; MSU claim UNCONFIRMED.
    • Chainalysis — External actor used "offensive agentic techniques" to reach internal systems and tools; user IDs, names, emails and role labels exposed. Attacker/method: Unnamed; AI agent-driven intrusion. Confidence: Confirmed by victim.
    • Frontline Education (K-12 SaaS) — 1,298,917 people in five state AG filings; SSNs, licence and Medicaid IDs. Breach in Aug; new totals 3 Nov. Attacker/method: Unnamed; exploited a third-party software flaw. Confidence: SINGLE SOURCE reporting of AG filings.

    Also confirmed this week: Florida Department of Education, about 4.9M people potentially affected (official notice). Mercado Bitcoin (Brazil) halted withdrawals after a theft; loss CONFLICTING (>$46M vs >$40M) and not confirmed by the exchange.

    Expand Post
  • SamHouston (Veracode)

    3. Threat Landscape (7 and 30 days)

    Ransomware

    • Synology's 29 Oct warning (Action 4) is the main new campaign. A claim that it uses PHP2Shell or GitLab flaws is UNCONFIRMED and contradicted by Synology's own advisory.
    • CISA/FBI advisory AA26-293A (20 Oct) says SafePay has hit at least 600 organizations since August 2024.
    • October leak-site counts are CONFLICTING: 1,127 claims (HackerFeeds; N0n 154, The Gentlemen 143, Qilin 140) vs 953 (Breachsense). Comparitech counts only 36 confirmed incidents.
    • Secondhand: an alleged core Qilin member was extradited from Japan to Germany on 2 Oct; Qilin still ranked third in October.

    AI-run intrusions

    • Sophos (29 Oct, original research): Aur0ra ransomware operators (GOLD OGRE) gave Cursor-hosted AI agents credentials and goals. The agents ran AD discovery and lateral movement with limited human input and reached domain admin in some environments. Guardrails were bypassed by framing the work as "red team" testing.
    • Microsoft says China-aligned Storm-0940 has used the ARTEX agentic framework since September. North Korea's Onyx Sleet and Pearl Sleet have used it since late October, sometimes reaching privileged access in under 3 hours.
    • CrowdStrike reports China-nexus VAPOR PANDA using ARTEX against 200+ organizations. Compromise counts are CONFLICTING: at least 9 confirmed (CrowdStrike) vs about 140 (CyberScoop). Use 9.
    • Chainalysis confirmed an agentic intrusion (Section 2).
    • UK NCSC Annual Review (3 Nov): 80 nationally significant incidents. AI is making attacks "faster, cheaper and easier" but has not created a new class of attack.
    • Gap: no new primary reports this week on prompt-injection attacks against MCP servers or coding agents.

    Identity

    • No public update since 21 Oct on the ConnectWise Cloud Backup app-credential abuse (Storm-0940). The attackers signed in as the app, so user MFA did not apply. Victim-tenant counts CONFLICT (173 vs 162).
    • Okta Access Gateway is affected by PHP2Shell; separately, CVE-2026-115547 is fixed in OAG 2026.10.2.
    • SIGNAL (single source, Dataminr): Okta- and ServiceNow-themed vishing that pushes YubiKey users to fall back to SMS codes.
    • The Ivanti attackers went after LAPS passwords and BitLocker keys (Action 1).

    Nation-state

    • North Korea: FBI attribution of the Binance theft (Section 2). Microsoft (29 Oct) reports DPRK AI use across operations: Jade Sleet ran fake-recruiter lures through three front companies in one week; Coral Sleet built a capability to trojanize private source repositories; Jasper Sleet builds fake IT-worker personas. Microsoft recommends blocking device-code sign-in.
    • China: a joint FBI/CISA/NSA advisory (8 Oct) describes automated and hands-on tooling.
    • Russia: ESET reports (22 Oct, covering a 2025 attack) that UAC-0099 handed access to Sandworm, which wiped a Ukrainian grain firm, using the Yggdrasil network for command and control.
    • Iran (SIGNAL, secondhand): "DarkBlinders" used a fake "StarkMeet" meeting app against Kurdistan Region government and Israeli targets.
    • Supermarket refrigeration incidents across five European countries (2 Nov) are not attributed.

    Cloud and infrastructure

    • Edge devices remain the main entry point (Ivanti).
    • Barracuda (2 Nov) disclosed CVE-2026-116606 and CVE-2026-111560, which chain to root on ESG, WAF and other appliances. They need admin-interface access, and no exploitation has been reported.
    • SINGLE SOURCE: the update channel of cPGuard, a Linux hosting security tool, was hijacked to push malicious versions 5.88.12-5.88.19; the fix is 5.88.25+.
    • GitHub Actions abuse continues: GitGuardian counts 72,026 GhostAction commits to 22,964 repositories (13-21 Oct; single source). There has been no public update on Gheist since 25 Oct.
    Expand Post
  • SamHouston (Veracode)

    4. Malicious Libraries and Components

    • @goparrot-dashboard/*, @goparrot-dashboard-api/* (25 pkgs, e.g. /store 0.0.1, /app 0.0.2) NEW 1-3 Nov (npm) — Dependency-confusion probe next to the real @goparrot scope; /app 0.0.2 preinstall sends host, user and directory to an oast.fun callback (MAL-2026-18511). Action: Block scopes; pin internal scopes to internal registry; search DNS/proxy logs for oast.fun.
    • testmetest worm (*-testmetest, @iobandamwer/*, @testsecuiitytestme/cache-helper) - re-checked (npm) — Steals SSH, registry, cloud and AI-tool credentials; self-publishes; plants Claude Code/Gemini/Cursor/VS Code hooks; wipes home directory if its token is revoked. No compromises reported beyond JFrog's 42 armed repositories. Action: Remove persistence before revoking tokens; --ignore-scripts does not protect (payload runs on import); audit for .github/init.js.
    • tensorlake 0.5.144 (8 Oct) (npm) — Hijacked release pipeline; credential-stealing worm with dead-man wiper. Action: Remove; isolate host, then rotate.
    • 106 packages flagged 26-28 Oct (e.g. dsh-plugin-blender-3d@0.2.1, @walknpm/todo-cli@2.0.1) - SECONDHAND summary (npm, PyPI) — Monero miners; dependency-confusion beacons (@greenlake, @sxm-web and others). Action: Search install logs from 23 Oct.
    • skills-analyzer 0.1.0-0.1.2 + is-it-ai NEW 31 Oct-3 Nov (PyPI) — Fingerprints host, downloads and runs a remote executable (MAL-2026-18427). Action: Block; check transitive dependencies.
    • pystr-codec, claudeopuseni, claudeopusenifile, claudeopusfile, zagents-framework, zagents-message-bus, nebulallm, poc-revshell, log-utils-sdk (PyPI) — Malware records published in the last 7 days (entries not individually analyzed). Action: Block names.
    • litellm 1.107.0 (PyPI and Docker, 21 Oct) (PyPI) — Release pipeline compromised; credential stealer via Python startup hooks (cook.pth, sitecustomize.py). Action: Pin 1.106.2; hunt for the files; rotate secrets on Python hosts.
    • dopc-client 1.0.0, 10.0.0; shoaib-unity; intercom-react-native-example-e2e-tests; macs-shoaibs-pro17 NEW 2-3 Nov (RubyGems) — Beacon on load to 38.242.235[.]206 (MAL-2026-18518). Action: Block.
    • ~24 crypto-themed gems at 0.9.1 (e.g. web3-signer-kit, eth-gas-oracle) (11 Oct) (RubyGems) — Install-time reverse shell; skips CI hosts (MAL-2026-17801). Action: Block; check developer machines.
    • sharpnes 0.1.0, 0.1.1 (9 Oct) (crates.io) — Steals Telegram sessions and browser wallet data. Action: Block.
    • anthropic.claude-code 2.1.312 and 2.1.313 (Cursor marketplace only) (IDE extension) — SSH backdoor (svc-telemetry key), reverse tunnel, persistence. Users affected: "at most 5,000" was dropped from Cursor's revised notice (CONFLICTING). Action: See Action 2; allowlist extensions.
    • "Searchjacked": 319 extensions, ~13.4M installs (29 Oct) (Chrome) — Search hijacking that can be switched to phishing. Action: Inventory and remove via browser management.
    • Fake Phantom, Jupiter and Bitget wallets (27 Oct) (Firefox) — Steal recovery phrases and private keys. Action: Block; brief crypto-handling staff.
    • "Load/Mine": 31 repos incl. MENXIANGYU/qwen2-tiny, 32,000+ downloads (29 Oct) (Hugging Face models) — Abuses trust_remote_code to install a miner; some variants open a reverse shell or attempt container escape. Action: Disallow trust_remote_code by default; scan models.
    • SolanaUpdateTools/solana-starter-kit - SINGLE SOURCE (GitHub / AI agents) — Repository built to lure AI coding agents into running a RAT. Action: Restrict agent auto-execution.

    No new public reports: Go modules, Maven Central, VS Code Marketplace and Open VSX, JetBrains Marketplace, and Edge Add-ons (7 and 30 days). crates.io has had nothing new in 7 days. For NuGet, Composer/Packagist and MCP servers/agent skills we found nothing, but coverage was incomplete, so treat this as UNCONFIRMED rather than clean.

    Expand Post
  • SamHouston (Veracode)

    5. Critical Vulnerabilities

    • CVE-2026-111357, CVE-2026-118527 — Ivanti Connect Secure. Exploited zero-days chained with CVE-2026-104935 (KEV 13 Oct) for unauthenticated root. Severity CONFLICTING (Ivanti 8.8 vs NVD/MITRE 7.0). Fix/deadline: 25.1.1.3 / 22.8R2.6. KEV added 30 Oct; due 2 Nov (passed) / 13 Nov.
    • CVE-2026-97229 — GitLab CE/EE. Unauthenticated RCE via crafted regex in CI/CD config; CVSS 10; Metasploit reference; attempted on Alpine Linux's GitLab; >1,000 exposed unpatched (SINGLE SOURCE). Fix/deadline: 19.2.8 / 19.3.5 / 19.4.2 (Ruby 3.3.13 / 3.4.12 / 4.0.8 for source installs). KEV 29 Oct; due 1 Nov (passed).
    • CVE-2026-111497 — Google Brotli (bundled in Node.js, .NET, Envoy, etc.). CISA lists an unchecked-return-value flaw in the decoder; Imperva showed code execution on a Node.js image; wide bundled exposure. Fix/deadline: Brotli 1.0.10 / 1.1.1 / 1.2.1; Node 22.23.4 / 24.22.1 / 26.12.1; .NET 8.0.33 / 9.0.22 / 10.0.14. KEV 27 Oct; due 10 Nov.
    • CVE-2026-48356 — Adobe Commerce / Magento. Dangerous file upload; patched in July (APSB26-73). Fix/deadline: KEV 27 Oct; due 10 Nov.
    • CVE-2026-41709 — VMware ESX. Insufficient logging; fixed version not confirmed. Fix/deadline: KEV 28 Oct; due 11 Nov.
    • CVE-2026-76418 — Cisco Secure Firewall ASA / FTD. Unauthenticated reload (DoS) via SSL VPN when secondary auth is set; listed as exploited by CISA. Fix/deadline: Fixed releases per NVD (no fix listed for FTD 7.6.x). KEV 29 Oct; due 1 Nov (passed).
    • CVE-2026-109618, CVE-2026-109619 + CVE-2026-46300 — PHP ("PHP2Shell") + Linux kernel. Carried over, no material change. Unit 42 retracted one earlier victim claim (SINGLE SOURCE). Fix/deadline: PHP 8.5.12 / 8.4.27 / 8.3.36 / 8.2.35. KEV 26 Oct (due 29 Oct) and 28 Oct (due 31 Oct).
    • CVE-2026-110936; CVE-2026-96652, CVE-2026-96655 — Siemens Software Center; Plex Media Server. Exploited; due today. Fix/deadline: KEV 20 Oct; due 3 Nov.
    • CVE-2026-102764 — BeyondTrust Remote Support. Probing seen 1 Nov from one IP, without success (SIGNAL). Fix/deadline: BT26-05-RS. KEV 16 Oct; due 19 Oct.
    • CVE-2026-116606 + CVE-2026-111560 — Barracuda ESG, WAF, ADC and others. Chain to root; needs admin-interface access; not exploited, not KEV. Fix/deadline: Barracuda advisory 2 Nov. ESG firmware CONFLICTING: 9.4.0.029 vs 9.4.0.030.
    • CVE-2024-10441 / -50629 / -10445 (not specified) — Synology DSM. Patched flaw exploited for ransomware. Fix/deadline: DSM 7.2.2-72806-1, 7.2.1-69057-6, 7.2-64570-4, 7.1.1-42962-7, 6.2.4-25556-8 or later.

    All of the 20 newest KEV entries show "Unknown" for ransomware use. We found no browser zero-days or Microsoft out-of-band patches between 27 Oct and 3 Nov. Android's November bulletin fixes a critical no-click Framework RCE (CVE-2026-58868); no exploitation is noted (secondhand).

    Expand Post
  • SamHouston (Veracode)

    6. Underground and Leak Signals (OSINT)

    Every claim below is UNCONFIRMED unless marked otherwise. Sources are published reporting only; no leak sites were visited.

    • BYOD: listed Michigan State University (1 Nov), claiming 1.3M student records.
    • N0n: listed 38 maritime and port organizations; none has confirmed a compromise.
    • The Gentlemen: claims MoneyGram, which has not confirmed.
    • Rhysida: offering 2.6TB said to be Anne Arundel County data. The county confirmed an attack, but the data's origin is unverified.
    • Forum sales (Bitsight, SIGNAL): "Z-Union" is selling data attributed to Parle Agro and NaRaYa; "OscarElMasCapito" is offering data from a Peruvian government agency.
    • Binance laundering: the FBI published 608 Ethereum, 19 Bitcoin and 1 Solana addresses to block (confirmed). Chainalysis, as reported, says more than $500M moved through 1,000+ TRON addresses, with funds shifting to Solana (single source).
    • Law enforcement (confirmed): Europol (3 Nov) seized 100+ servers behind investment-fraud sites and made 7 arrests; losses exceed €4M, with victims in 50+ countries.

     

    7. Foresight (next 2-6 weeks): ANALYST ASSESSMENT

    This section is judgment, not reporting.

    • Patch Tuesday, 10 Nov. October was reported as a record month (1,411 or 1,413 CVEs; sources CONFLICT), and the NCSC warns of an AI-driven "patch wave". Expect another heavy month; budget triage time in the same week as the 10-13 Nov KEV deadlines.
    • Agentic intrusions become routine. ARTEX now appears in reports on two Chinese and two North Korean groups, and Sophos expects more ransomware crews to adopt agents. Plan for privilege escalation in hours and more abuse of OAuth apps and service principals.
    • Developer-tool supply chain stays the softest target. In two weeks we have seen an IDE marketplace backdoor, worms that plant AI-assistant hooks, and daily dependency-confusion probes. These are likely to continue. GitHub's reported 2 Dec change to pull_request_target defaults is SINGLE SOURCE; confirm it on GitHub's changelog.
    • North Korean follow-on. After Binance, expect continued fake-recruiter and developer-laptop targeting of crypto, fintech and software firms.
    • Holiday fraud (Black Friday, 27 Nov). The FBI cites $503M+ in 2025 non-delivery and non-payment losses. Retail and payment-card skimming risk rises.
    • Policy dates: CISA BOD 26-04 timelines from about 7 Dec (secondhand). The US Cybersecurity Information Sharing Act of 2015 reportedly expires 11 Dec (SINGLE SOURCE).
    Expand Post
  • SamHouston (Veracode)

    8. CISO Control Checklist

    • Inventory Ivanti ICS; upgrade to 25.1.1.3 / 22.8R2.6; hunt off the appliance; rotate LDAP service credentials — Tied to: Ivanti chain (Action 1). Owner: Network Security + Incident Response.
    • Sweep endpoints for svc-telemetry key and Claude Code 2.1.312/2.1.313 from Cursor; enforce an IDE-extension allowlist — Tied to: Cursor marketplace backdoor (Action 2). Owner: Endpoint/IT + Developer Experience.
    • Map internal package scopes to the internal registry only; block new MAL- packages; audit lockfiles from 23 Oct — Tied to: goparrot, is-it-ai, dopc-client, testmetest (Action 3). Owner: AppSec + Platform Engineering.
    • For worm-infected hosts: remove persistence before revoking tokens — Tied to: testmetest, tensorlake. Owner: Incident Response.
    • Disallow trust_remote_code; scan models before use; restrict AI-agent auto-execution — Tied to: Hugging Face Load/Mine; agent-lure repos. Owner: ML Platform + AppSec.
    • Audit CI for ci-sync-* branches, unexpected workflow runs and over-scoped secrets; prefer OIDC short-lived credentials — Tied to: Gheist, GhostAction, LiteLLM. Owner: DevOps / Platform Security.
    • Review OAuth apps and service-principal secrets; block device-code flow; alert on app-only sign-ins — Tied to: Storm-0940 / ARTEX, ConnectWise. Owner: Identity & Access Management.
    • Separate developer workstations from production signing and payment systems; brief staff on fake-recruiter lures — Tied to: Binance / TraderTraitor; DPRK AI lures. Owner: Security Engineering + Talent/HR comms.
    • Find all Synology NAS; update DSM; remove internet-facing management; keep offline backups — Tied to: C3RB3R (Action 4). Owner: IT Operations.
    • Track KEV deadlines: GitLab (correct versions), Brotli images, Adobe Commerce, VMware ESX — Tied to: KEV (Action 5). Owner: Vulnerability Management.
    • Check ransomware detections against SafePay advisory TTPs — Tied to: AA26-293A. Owner: Security Operations.

     

    Sources

    Publisher, title, date.

    • CISA - Known Exploited Vulnerabilities Catalog (read 3 Nov 2026); CISA Adds Two Known Exploited Vulnerabilities to Catalog (27 Oct 2026); AA26-293A #StopRansomware: SafePay Ransomware (20 Oct 2026)
    • NIST NVD - CVE-2026-97229 (modified 2 Nov 2026); CVE-2026-111357; CVE-2026-111497; CVE-2026-76418; CVE-2026-109618; CVE-2026-46300
    • FBI - North Korea Responsible for $6 Billion+ Binance Attack (31 Oct 2026); Holiday Scams
    • Ivanti - Security Advisory Ivanti Connect Secure (CVE-2026-104935) (updated 29 Oct 2026)
    • Cursor - Security notice: malicious version of Anthropic.claude-code (2.1.312 and 2.1.313) (29 Oct 2026, rev. 30 Oct)
    • Synology - Synology urges users to update DSM following ransomware attacks on outdated DSM versions (29 Oct 2026)
    • GitHub Advisory Database / OpenSSF - GHSA-726f-wf9g-5893 (3 Nov 2026); GHSA-r92r-m9r8-78vj; GHSA-8v7q-mrx4-jvfg (3 Nov 2026); GHSA-qq2r-xp7w-5r55
    • Barracuda - Pre-authentication authorization bypass and command injection advisory (2 Nov 2026)
    • Okta - Command execution via admin console input in Okta Access Gateway (CVE-2026-115547) (22 Oct 2026)
    • SEC EDGAR - Revolve Group Form 8-K (29 Oct 2026); Florida Dept. of Education - Data Breach notice (29 Oct 2026); UNM Health Sciences - cybersecurity issue notice (26 Oct 2026); Alpine Linux - GitLab security incident (Oct 2026); Europol - Investment scams... over 100 servers seized (3 Nov 2026); UK NCSC - Annual Review 2026 (3 Nov 2026)
    • Microsoft Security Blog - Inside a malicious Claude Code-branded extension... (30 Oct 2026); Tracking AI adoption in North Korean threat actor tradecraft (29 Oct 2026); Storm-0940 uses ARTEX... (21 Oct 2026); Inside the LiteLLM 1.107.0 supply chain compromise (23 Oct 2026)
    • Sophos - AI used in Aur0ra ransomware testing and execution (29 Oct 2026); CrowdStrike - VAPOR PANDA uses ARTEX to scale global campaign (29 Oct 2026); Volexity - Chained exploitation of Ivanti Connect Secure zero-days (13 Oct 2026); Wiz - Gheist (25 Oct 2026); JFrog - Beyond testmetest (26 Oct 2026); GitGuardian - GhostAction x PolinRider (27 Oct 2026); Pluto Security - Load/Mine (29 Oct 2026); Socket - Searchjacked (29 Oct 2026), Firefox wallet impersonators (27 Oct 2026), tensorlake compromise (8 Oct 2026); TRM Labs - North Korea steals USD 6.36 billion from Binance (29 Oct 2026); LiteLLM - v1.107.0 security incident (21 Oct 2026)
    • SecurityWeek - Ivanti Patches Two More Zero-Days... (2 Nov 2026); Malicious Claude Code Extension Served in Cursor Supply Chain Attack (2 Nov 2026); Revolve Data Breach Impacts at Least 180,000 People (2 Nov 2026); Frontline Education Data Breach Impacts Over 1.3 Million (3 Nov 2026)
    • BleepingComputer - Synology warns of C3RB3R ransomware attacks targeting NAS devices (2 Nov 2026); Ivanti patches two more Connect Secure zero-days (2 Nov 2026); Alpine Linux rebuilds GitLab after attempted RCE attack (1 Nov 2026); Barracuda warns of critical flaws... (3 Nov 2026); Chainalysis discloses data breach... (Oct 2026); UNM Health ransomware attack disrupts patient care (27 Oct 2026)
    • The Hacker News - Synology NAS Devices Targeted... (3 Nov 2026); FBI Links North Korea's TraderTraitor to Record $6.43 Billion Binance Crypto Heist (1 Nov 2026); Attackers Chain Three Ivanti Connect Secure Flaws (31 Oct 2026); cPGuard update channel hijacked (3 Nov 2026); DarkBlinders (2 Nov 2026)
    • Help Net Security - CISA to federal agencies: Remediate exploited Ivanti Connect Secure vulnerabilities (2 Nov 2026); Aur0ra ransomware AI agents (2 Nov 2026)
    • The Record - Michigan State University BYOD incident (2 Nov 2026); Mercado Bitcoin heist (2 Nov 2026); Sandworm grain-firm wiper (22 Oct 2026). CyberScoop, GovExec, CNBC - DHS staff leak (28-29 Oct 2026). CoinDesk - North Korea denies role (3 Nov 2026). Unchained - FBI wallet list (2 Nov 2026)
    • HackerFeeds, Breachsense, Comparitech - October 2026 ransomware statistics; Bitsight - underground data-breach feed; Dataminr - Okta-themed phishing intel brief (19 Oct 2026)

     

    This report is provided for informational purposes only and is not intended as legal, technical, or professional advice. While we strive for accuracy, Veracode does not warrant the completeness or accuracy of the information. Recipients should not rely solely on this report and must conduct their own thorough investigation and verification. Please work with your internal teams and relevant stakeholders to properly assess, implement, and remediate any identified threats or vulnerabilities. The information has been compiled from multiple sources, and Veracode assumes no liability for any errors, omissions, or actions taken based on this content.

    Expand Post

Topics (4)

No articles found
Loading

Ask the Community

Get answers, share a use case, discuss your favorite features, or get input from the community.