pwn.watch: all incidents updates

[high] Ahsay AhsayCBS: Ahsay AhsayCBS Command Injection Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Ahsay AhsayCBS contains a command injection vulnerability that could allow a remote attacker to execute operating system commands via manipulation of the random parameter in the Replication Receiver component’s /rps/api/json/UpdateReceivers.do endpoint. This vulnerability can be chained with CVE-2026-105133. CISA added CVE-2026-105134, CVE-2026-105133 to the Known Exploited Vulnerabilities catalog on 2026-10-14 based on evidence of active exploitation.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-14: CISA added CVE-2026-105134, CVE-2026-105133 to the KEV catalog (due 2026-10-17).

CVEs: CVE-2026-105134, CVE-2026-105133

[critical] Arista VeloCloud Orchestrator zero-day CVE-2026-114625 exploited in attacks

Fri, 23 Oct 2026 11:12:55 GMT

Arista patched an actively exploited critical SQL injection flaw (CVE-2026-114625, CVSS 9.9) in on-prem VeloCloud Orchestrator that lets any authenticated user gain operator privileges and masquerade as the local 'superuser operator'. Hosted VCO is already patched; on-prem admins need to upgrade.

Status: live · Patch: patched · Fixed in: 5.2.3.18, 6.1.3.10, 6.4.2.12

Latest updates:

2026-10-22: Arista disclosed CVE-2026-114625, an actively exploited SQL injection in on-prem VeloCloud Orchestrator, with fixes in 5.2.3.18, 6.1.3.10 and 6.4.2.12. Hosted VCO is already patched. Arista advised restricting the VCO web interface to admin networks, blocking 74.82.197.86 and 38.54.50.151, and preserving logs before remediating if compromise is suspected. (source)

CVEs: CVE-2026-114625

[high] Atlassian Multiple Products: Atlassian Multiple Products Files or Directories Accessible to External Parties Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Multiple Atlassian products (Bitbucket Data Center ,Confluence Data Center, Jira Service Management Data Center, Jira Software Data Center, Bamboo Data Center, Crowd Data Center, Crucible and Fisheye) contain a files or directories accessible to external parties vulnerability which could allow an unauthenticated attacker to access specific files within the web application root directory in affected versions. Added to CISA KEV on 2026-10-15.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-15: CISA added CVE-2026-21589 to the KEV catalog (due 2026-10-18).

CVEs: CVE-2026-21589

[critical] BeyondTrust Remote Support zero-day CVE-2026-102764 (ARMORY CASTLE)

Fri, 23 Oct 2026 11:12:55 GMT

China-nexus actor ARMORY CASTLE (ex-UNC5221) exploited CVE-2026-102764, a critical unauthenticated file-operations flaw in BeyondTrust Remote Support, as a zero-day from late August 2026. BeyondTrust issued BT26-05-RS; SaaS was auto-patched by Oct 9; CISA added the CVE to KEV on Oct 16. Self-hosted appliances need the patch plus forensics — patching alone does not clear existing footholds (SAWFLY / BARECLAW).

Status: live · Patch: partial · Fixed in: BT26-05-RS patch for self-hosted RS 24.3.1-26.2.3 (older versions must upgrade first); SaaS auto-patched Oct 9

Latest updates:

2026-10-15/16: Public disclosure / Mandiant blog; CISA KEV Oct 16.
2026-10-08/09: BeyondTrust patches; SaaS auto-fixed.
2026-08-29: Mandiant/GTIG: exploitation of zero-day begins (limited victims).

CVEs: CVE-2026-102764

[high] Capcom ID breach exposes 11.7M account email addresses

Fri, 23 Oct 2026 11:12:55 GMT

Capcom says a third party gained unauthorized access to CAPCOM ID, the account needed to play some Capcom games online. 11,774,247 records worldwide had email addresses plus account creation time, last login time, login count and last-login IP exposed; 43 also had date of birth, gender, country/region and nickname. Passwords were changed without authorization on 196 accounts (already reset). Capcom says it has not confirmed any leak of customer-set passwords, and no credit card data was involved. Detected Oct 19; reported to Osaka Prefectural Police and Japan's PPC on Oct 21. Expect phishing impersonating Capcom.

Status: live · Patch: patched · Fixed in: Capcom says emergency measures are complete and the attack method was identified and addressed; services remain available

Latest updates:

2026-10-22: Public disclosure: 11,774,247 records (email, account creation and last-login times, login count, last-login IP), 43 with date of birth, gender, country and nickname, 196 accounts with unauthorized password changes (reset). No credit card data. Affected customers will be notified. (source)
2026-10-21: Capcom reports the incident to the Osaka Prefectural Police and Japan's Personal Information Protection Commission. (source)
2026-10-20: Capcom sets up an internal incident response headquarters and confirms customer data was accessed and some account passwords were changed. (source)

[high] Cisco Secure Email Gateway: Cisco Secure Email Gateway Relative Path Traversal Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Cisco Secure Email Gateway contains a relative path traversal vulnerability that could allow an unauthenticated, remote attacker to write arbitrary files on the underlying operating system. Added to CISA KEV on 2026-10-14.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-14: CISA added CVE-2026-76519 to the KEV catalog (due 2026-10-17).

CVEs: CVE-2026-76519

[high] Citrix NetScaler: Citrix NetScaler Improper Restriction of Operations within the Bounds of a Memory Buffer Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Citrix NetScaler ADC (formerly Citrix ADC) and Citrix NetScaler Gateway (formerly Citrix Gateway) contain an improper restriction of operations within the bounds of a memory buffer vulnerability that could allow for a denial of service. Added to CISA KEV on 2026-10-04.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-04: CISA added CVE-2026-88779 to the KEV catalog (due 2026-10-07).

CVEs: CVE-2026-88779

[critical] ConnectWise Cloud Backup shut down after M365 app-identity hijack

Fri, 23 Oct 2026 11:12:55 GMT

ConnectWise disabled Cloud Backup after a compromised credential let attackers authenticate as the application and sign in to at least 173 customer Microsoft 365 tenants.

Status: live · Patch: partial

Latest updates:

2026-10-22: Microsoft details Storm-0940's Oct 10-14 password spray (151,000+ failed logons, peak 2,900/min) before the Oct 15 takeover of ConnectWise cloud app identities; ConnectWise says phased service restoration will take several days. (source)
2026-10-22: ConnectWise CEO Manny Rivelo says the affected application connections are contained and services are being restored gradually, but cutting those connections won't necessarily evict attackers who already set up persistent access in customer tenants. Microsoft hasn't said how many customers were compromised. (source)
2026-10-22: Wiz Research says the same ARTEX-using actor, overlapping with Storm-0940, is also targeting other IT and software vendors: password-spraying Okta and Entra, abusing Cognito and account-recovery endpoints, and probing Microsoft 365 integrations and tenant isolation. Wiz published IPs and user agents to hunt for. (source)

[high] Hijacked website tags and Instagram ads deliver Coruna and DarkSword spyware to outdated iPhones

Fri, 23 Oct 2026 11:12:55 GMT

Report URI and iVerify found criminals using 660+ re-registered expired analytics and ad-tag domains (DeadTag), hacked PrestaShop stores and a malicious Instagram ad to silently serve the Coruna (iOS 13 to 17.2.1) and P7 DarkSword (iOS 18.4 to 18.7) exploit kits. Opening an affected page on a vulnerable iPhone is enough; the spyware steals keychain passwords, crypto wallet data and device info. Updated iPhones are protected, so update iOS (or turn on Lockdown Mode if you can't).

Status: live · Patch: patched · Fixed in: Latest iOS 26, Latest updates for iOS 15 through iOS 18 (Apple says updated devices are protected)

Latest updates:

2026-10-22: Report URI detailed the DeadTag operation: 660+ re-registered expired tag domains still loaded on thousands of sites (an estimated 135 million visits a month), selling visitor traffic to buyers including a Chinese-speaking crew that serves Coruna and DarkSword from a fake Polymarket page. The crew's DarkSword loader was also found on hundreds of hacked PrestaShop stores. (source)
2026-10-21: iVerify and Report URI published joint research: a World Cup themed Instagram ad served from ad-track[.]live infected a financial-institution employee's iPhone with P7 DarkSword while they scrolled their feed, and expired analytics and ad-tag domains were redirecting iPhone visitors to Coruna and DarkSword. (source)

[high] EG Secure Solutions SiteGuard Server Edition: EG Secure Solutions SiteGuard Server Edition OS Command Injection Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

EG Secure Solutions SiteGuard Server Edition contains an OS command injection vulnerability which may allow an attacker to execute arbitrary OS commands with root or SYSTEM privileges on the system running the affected product. Added to CISA KEV on 2026-10-20.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-20: CISA added CVE-2026-106931 to the KEV catalog (due 2026-10-23).

CVEs: CVE-2026-106931

[critical] eMudhra intermediate CA private keys compromised

Fri, 23 Oct 2026 11:12:55 GMT

Eight eMudhra Global TLS intermediate CA private keys were found to be cryptographically weak / recoverable (researcher report Oct 18). Mozilla, Chrome, Apple and Microsoft blocked the intermediates; eMudhra revoked them Oct 19 and the ECC root was distrusted as a precaution. Downstream runtimes that ship their own trust stores (Go, Node.js, Electron, Python certifi, curl, etc.) needed emergency CA-bundle updates so they stop trusting the compromised chains.

Status: live · Patch: partial · Fixed in: Electron 44.9.1 / 43.8.1 / 42.11.16, Go 1.27.3 / 1.26.10, Node.js 22.23.4 / 24.22.1 / 26.12.1, Python certifi 2026.10.19, NSS 3.131.1, Java 27.0.1.0.2 / 25.0.5.0.2 / 21.0.13.0.2 / 17.0.21.0.2 / 11.0.33.0.2 / 8u511+ (CVE-2026-91403), Deno 2.9.8, rustls-webpki 0.103.16 / webpki-roots 1.0.10, pip 26.2.2 / uv 0.13.4 / sigstore 4.5.1 / conda 26.9.4 / pixi 0.82.1 (Python ecosystem), Ubuntu/Debian/Alpine/Fedora/Arch ca-certificates updates (see tracker), Docker Desktop 4.96.2, mitmproxy 12.3.0, LibreSSL 4.2.3 / 4.3.4 / 4.4.0

Latest updates:

2026-10-22: Tracker lists patched releases across distros, language runtimes and tools; Bun, Erlang certifi and Bottlerocket still pending. mitmproxy 12.3.0 drops eMudhra CAs. (source)
2026-10-22: eMudhra says a flaw in its April 30 2025 key generation ceremony produced a second set of weak keys, and those were the ones published to root stores; root cause still under investigation, replacement certs being issued to customers. (source)
2026-10-22: Internet Freedom Foundation says it has written to eMudhra, India’s Controller of Certifying Authorities and CERT-In asking for a fix. (source)

CVEs: CVE-2026-111943, CVE-2026-91403

[critical] Fearless Wallet browser extension leaked private keys for ~20 months

Fri, 23 Oct 2026 11:12:55 GMT

The Polkadot Community Foundation warned that Fearless Wallet browser-extension versions 3.0.1, 3.0.3, 3.0.4 and 3.0.5 contained code, added Feb 19 2025 inside a large 'TON ecosystem' change, that sent private keys/recovery data to an attacker server when the wallet was unlocked. The code was active at different times until Oct 10 2026. Eight users have reported losses totalling 17,286 DOT. Anyone who used the extension since Feb 2025 must create a new keypair in another wallet and move funds.

Status: live · Patch: unpatched

Latest updates:

2026-10-21: Polkadot Community Foundation publishes security notice (14:00 UTC); 8 users report 17,286 DOT lost. (source)
2026-10-10: Last known period the exfiltration code was active (it could be re-enabled). (source)
2025-02-19: Malicious code merged into the extension's public source as part of a 'TON ecosystem' PR. (source)

[critical] FortiMail unauthenticated file-write zero-day CVE-2026-104286

Fri, 23 Oct 2026 11:12:55 GMT

Fortinet FortiMail CVE-2026-104286 (CVSS 9.8) is an unauthenticated arbitrary file write via path traversal / null-byte issues in Identity-Based Encryption web flows. CISA confirmed active exploitation and ordered federal mitigation; Fortinet published FG-IR-26-175 with patches and interim mitigations (disable IBE / restrict admin/webmail).

Status: live · Patch: patched · Fixed in: 8.0.2+, 7.6.7+, 7.4.9+; 7.2.x migrate to 7.4.9+

Latest updates:

2026-10-05: Fortinet remediates FortiMail Cloud by updating all instances to 7.6.7 / 8.0.2 GA. (source)
2026-10: Fortinet PSIRT FG-IR-26-175 patches and mitigations.
~2026-09-30/10-01: Active exploitation and CISA alerts begin circulating.

CVEs: CVE-2026-104286

[high] Fortinet Multiple Products: Fortinet Multiple Products Path Traversal Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

FortiOS, FortiPAM, FortiProxy, and FortiSwitchManager contain a path traversal vulnerability that may allow an authenticated attacker with admin profile and at least read-write permissions to write or delete arbitrary files via specific CLI commands. Added to CISA KEV on 2026-10-16.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-16: CISA added CVE-2025-61624 to the KEV catalog (due 2026-10-19).

CVEs: CVE-2025-61624

[critical] Hackers hijack refrigeration controllers at about 700 French supermarkets (Danfoss, Copeland and Dixell XWEB)

Fri, 23 Oct 2026 11:12:55 GMT

Overnight on Oct 11 to 12, attackers took remote control of internet-exposed refrigeration controllers at about 700 French supermarkets (Intermarché and Netto, Coopérative U, Carrefour with 80 stores, Monoprix, E. Leclerc and others), forcing fridges and freezers into continuous defrost with setpoints around 50°C, spoiling fresh and frozen stock and reportedly causing small fires. A second wave followed the next night and also hit stores with Dixell controllers. CERT-FR names Danfoss AK-SM and AK-SC series plus Copeland and Dixell xweb controllers, says campaigns are probably still ongoing, and tells operators to take every controller off the public internet. Most affected Danfoss units are end-of-life models (AK-SM 800, AK-SC 355) that get no fixes; the entry vector and attacker are still unconfirmed.

Status: live · Patch: partial · Fixed in: No fix for end-of-life units: remove internet exposure and replace them, Danfoss AK-SM 800A: update to current firmware, security level Strict, change default passwords (CERT-FR)

Latest updates:

2026-10-22: Dark Reading reports a little over 700 stores were hit in one night; a Sekoia analyst told Reuters Spain, Portugal, Poland, Germany and Belgium were also affected, and Les Mousquetaires (Intermarché, Netto) said the attack came through a refrigeration service provider. (source)
2026-10-16: CERT-FR publishes a third version with separate steps for compromised and uncompromised units: remove internet exposure, check defrost ranges, restore access only via VPN, and reset and reconfigure compromised Danfoss units. (source)
2026-10-15: France's finance minister has told lawmakers about 700 of the country's roughly 10,000 supermarkets were affected; Copeland now acknowledges a vulnerability and says it is working with impacted customers. Censys saw exposed Danfoss StoreView panels in France drop from 1,800+ to 537 in three days. (source)

[high] GhostAction returns: hijacked GitHub accounts push secret-stealing 'security audit' workflows to 30k+ repos

Fri, 23 Oct 2026 11:12:55 GMT

GhostAction, a campaign first documented by GitGuardian in Sept 2025, came back in 2026. It hit 772 repos in Aug-Sep 2026, then surged on Oct 8-12: about 1,900 compromised GitHub accounts pushed fake 'security audit' workflows to 30,000+ repositories (GitGuardian). The new payload exfiltrates the repo's Actions secrets plus every credential found in the full git history to 193.32.204.199 over plain HTTP. GitHub began preventing the workflow from running around Oct 9. The operator moved on to GitHub-themed phishing (my-github.com) and a VS Code tasks.json + fake .woff2 font stealer found in 94 repos.

Status: live · Patch: partial

Latest updates:

2026-10-16: OpenSourceMalware finds a second stage in 94 repos: VS Code tasks.json autoruns a stealer hidden in a fake Font Awesome .woff2 and installs a 5-minute cron. (source)
2026-10-13: GitGuardian update: Oct 8-12 wave targeted 30,000+ repos across ~3,000 users/orgs from ~1,900 compromised accounts; 3,174 secrets targeted. (source)
2026-10-11: Phishing kit on my-github.com steals password + 2FA and silently authorizes an OAuth app with repo/workflow scopes. (source)

[critical] GitLab CVSS 10 unauthenticated RCE via regex compiler (CVE-2026-110264)

Fri, 23 Oct 2026 11:12:55 GMT

GitLab released a critical patch on Oct 21 2026 for CVE-2026-110264: an integer overflow in the bundled Onigmo / regex compiler that can allow an unauthenticated attacker to execute code on self-managed GitLab. CVSS 10.0; every CE/EE version from 7.0 (2014) is in range. Instances that only applied the Oct 12 regex fix remain vulnerable and must upgrade again.

Status: live · Patch: patched · Fixed in: 19.5.1, 19.4.3, 19.3.6, 19.2.9, 19.1.9, 19.0.10, 18.11.13

Latest updates:

2026-10-22: Shadowserver tracks 2,700+ internet-exposed GitLab CE instances (honeypot and patch status unknown); researcher joaxcar reported the bug via HackerOne. (source)
2026-10-22: GitLab updated its advisory with two threat detections ("Onigmo native worker crash" and "Org mode render error burst") to help self-managed customers spot exploitation attempts; GitLab.com and Dedicated remain patched. (source)
2026-10-21: GitLab released 19.5.1 and backports for CVE-2026-110264; GitLab.com already patched.

CVEs: CVE-2026-110264

[medium] GoDaddy mass-revokes 404,607 TLS certificates over CP/CPS profile mismatch

Fri, 23 Oct 2026 11:12:55 GMT

GoDaddy (Starfield) found that 404,607 TLS certificates carried policy information (policy OIDs, CPS URI) that didn't match its own published CP/CPS, which under the CA/Browser Forum Baseline Requirements means they must be revoked within five days. GoDaddy missed the Oct 18 deadline and says the certificates were revoked on Oct 21 2026. Nothing was hacked and no keys leaked, but sites still serving a revoked certificate can show browser warnings or become unreachable until the re-issued certificate is installed.

Status: live · Patch: partial · Fixed in: Re-issued certificates from the GoDaddy SSL dashboard (customer must install them)

Latest updates:

2026-10-21: GoDaddy's customer notice says the incorrectly issued certificates were revoked; affected sites must install the re-issued certificate. (source)
2026-10-19: 227,422 of 404,607 certificates revoked as of 05:29 UTC. (source)
2026-10-18: Five-day revocation deadline (23:45 UTC) passes with 170,566 certificates revoked; GoDaddy files a delayed-revocation incident. (source)

[unknown] HTTP/3 QPACK “Header Accordions” memory-amplification DoS

Fri, 23 Oct 2026 11:12:55 GMT

FoxIO published research (Oct 20 2026) showing that missing or late decoded-header limits in QPACK let a tiny HTTP/3 HEADERS frame expand into huge server memory use (up to ~4000× on Bun). Seven of ten implementations tested were vulnerable; quic-go, Cloudflare quiche, lsquic and Wireshark are patched, while Bun, aioquic/mitmproxy and h3 remained unpatched at publication.

Status: live · Patch: partial · Fixed in: quic-go 0.57.0, Cloudflare quiche 0.29.3, lsquic 4.9.0, Wireshark 4.6.2 / 4.4.12

Latest updates:

2026-10-20: FoxIO blog + PoC published; several stacks still unpatched.
2025-11 → 2026-07: Responsible reports to quic-go, quiche, Bun, aioquic, lsquic, h3.

CVEs: CVE-2025-64702, CVE-2026-12523, CVE-2025-13945

[high] ISC BIND 9: replayed GSS-TSIG messages corrupt memory, possible RCE (CVE-2026-94300)

Fri, 23 Oct 2026 11:12:55 GMT

ISC disclosed a remotely exploitable memory-corruption bug in BIND 9's named on Oct 21. An attacker who captures a GSS-TSIG signed UDP message (the Kerberos signing used for Active Directory-style dynamic DNS updates) can replay it while the negotiated key is still valid and, combined with crafted queries to a zone they control, may get remote code execution, or at least crash named. CVSS 8.1 (High). No workaround. Affects BIND 9.11.0 through 9.18.50, 9.20.0 through 9.20.29 and 9.21.0 through 9.21.26. Fixed in 9.20.30 and 9.21.27; ISC lists no 9.18 fix. ISC knows of no active exploitation. Debian trixie has a fixed package; Debian bookworm and Ubuntu are still waiting.

Status: live · Patch: partial · Fixed in: BIND 9.20.30, BIND 9.21.27, BIND Supported Preview Edition 9.20.30-S1, Debian trixie: bind9 1:9.20.30-1~deb13u1 (DSA-6576-1); unstable 1:9.20.30-1

Latest updates:

2026-10-22: Debian publishes DSA-6576-1 with a fixed bind9 for trixie; bookworm (9.18) is still listed as vulnerable. (source)
2026-10-22: Japan's registry JPRS issues an urgent alert telling BIND operators to upgrade. (source)
2026-10-21: ISC publicly discloses CVE-2026-94300 and ships BIND 9.20.30 and 9.21.27. ISC found it in internal testing; Dave Aitel of OpenAI reported it independently. (source)

CVEs: CVE-2026-94300

[high] Ivanti Connect Secure zero-day chain (CVE-2026-104935) exploited for root

Fri, 23 Oct 2026 11:12:55 GMT

Volexity observed threat actor UTA0576 chaining an Ivanti Connect Secure authentication-bypass / RCE zero-day (CVE-2026-104935) with additional flaws from about Oct 1–3 2026 to get unauthenticated root on fully patched appliances, re-exploiting in memory without planting webshells. Ivanti patched CVE-2026-104935 on Oct 13; CISA added it to KEV. One privilege-escalation piece of the chain was still being fixed at disclosure.

Status: live · Patch: partial · Fixed in: 25.1.1.2, 25.1.3.0 / 25.1.3.1, 22.8R2.5

Latest updates:

2026-10-13: Ivanti discloses and patches CVE-2026-104935; Volexity publishes.
2026-10-13/16: CISA KEV listing; federal patching deadline reported as Oct 16.
2026-10-03: Volexity sees UTA0576 chaining for unauth root.

CVEs: CVE-2026-104935

[critical] Keycloak unauthenticated account takeover gets a public exploit (CVE-2026-18963)

Fri, 23 Oct 2026 11:12:55 GMT

A flaw in Keycloak's reset-credentials flow (keycloak-services) lets an unauthenticated attacker force a password reset for any user without clicking the emailed verification link, then set new credentials and take over the account. Red Hat rates it CVSS 9.1. Fixes shipped in August (26.4.15, 26.6.6, 26.7.2); on Oct 21 a full proof-of-concept with check and takeover modes was published on Exploit-DB, so unpatched Keycloak instances should be upgraded now.

Status: live · Patch: patched · Fixed in: 26.4.15, 26.6.6, 26.7.2, 26.8.0 and later

Latest updates:

2026-10-21: Exploit-DB publishes EDB-52699 "Keycloak 26.7.2 - Unauthenticated Account Takeover" by Abel Maldonado (Orevim): a PoC with a safe --check mode and a --set-password takeover mode, tested on Keycloak 26.6.2. (source)
2026-08-18: CVE-2026-18963 published with GitHub advisory GHSA-4gv3-mc9p-5wqc (critical); fixed in keycloak-services 26.4.15, 26.6.6 and 26.7.2, with Red Hat errata for Red Hat Build of Keycloak. (source)

CVEs: CVE-2026-18963

[critical] Linux kernel: unpatched ZDI bug reports with repro details leaked to public LKML (ksmbd RCE, CVSS 9.0)

Fri, 23 Oct 2026 11:12:55 GMT

On Oct 21, Trend Micro's Zero Day Initiative sent several unpatched Linux kernel vulnerability reports to security@kernel.org but Cc'd the public linux-kernel and subsystem lists, so full technical details (code paths, KASAN traces) are now public before fixes ship. The worst are two ksmbd (in-kernel SMB server) remote code execution bugs that ZDI scores CVSS 9.0: an SMB Direct integer underflow (needs an RDMA device and a writable share) and a protocol-dispatch race (needs a guest share). Also leaked: an skb_zerocopy local privilege escalation (CVSS 7.8), a vmwgfx out-of-bounds write LPE, and an nvmet info leak. No CVEs yet and no fixed mainline or stable release; the ksmbd race fix sits in the ksmbd maintainer tree and an SMB Direct fix is in a branch aimed at Linux 7.5. No exploitation reported. Only machines running ksmbd are exposed to the RCEs.

Status: live · Patch: unpatched · Fixed in: No fixed mainline or stable kernel yet, ksmbd race (ZDI-CAN-33774): fix commit 68f410faedaf in the ksmbd maintainer tree (linkinjeon/smb.git), not yet in a release, SMB Direct underflow (ZDI-CAN-33549): likely fixed in Stefan Metzmacher's for-7.5/smbdirect branch (unconfirmed)

Latest updates:

2026-10-22: Greg Kroah-Hartman replies to the reports: "For some reason you sent this to public lists, and did not include a patch" and "Please fix your submission process..." (source)
2026-10-22: Eric Dumazet says a patch for the skb_zerocopy bug, reported by someone else a week earlier, will be available soon. (source)
2026-10-22: Phoronix reports the leak. ksmbd maintainer Namjae Jeon says the ksmbd race is already fixed in his tree. (source)

[critical] LiteLLM 1.107.0 PyPI/Docker supply-chain compromise

Fri, 23 Oct 2026 11:12:55 GMT

On Oct 21 2026 attackers compromised LiteLLM’s release pipeline and published malicious litellm==1.107.0 to PyPI (live ~2 hours) plus compromised Docker Hub/GHCR tags (v1.107.0, latest, main-latest, main-stable). The wheels install cook.pth / cook.py / sitecustomize.py that steal cloud, developer and AI-tool credentials at every Python startup and open a remote-code loader. Second LiteLLM compromise of 2026.

Status: live · Patch: patched · Fixed in: 1.106.2 (pin; last known safe). Malicious 1.107.0 removed from PyPI, Docker Hub and GHCR

Latest updates:

2026-10-23: SecurityWeek: the malicious 1.107.0 packages were online for roughly three hours and are estimated to have been downloaded over 49,000 times; enterprise customers pulling from LiteLLM’s private repository were not affected. (source)
2026-10-22: LiteLLM root cause: attacker held three developer GitHub CLI tokens, likely scraped from logs of an internet-exposed internal LiteLLM deployment (Claude Code can echo a gh OAuth token into logged response bodies), disabled branch protection and ran the release workflow against lookalike repo Berri-AI/litellm. (source)
2026-10-22: StepSecurity full teardown: payload exfiltrates to lookalike www.litellm.dev (registered Oct 13) and an hourly loader polls a GitHub dead drop (letters9813/pythonclient-test, now deleted). Infected hosts keep polling, so anyone re-registering that username could run code on them: kill the process holding /tmp/remote_code_loader.lock, delete cook.pth/cook.py/sitecustomize.py, pin 1.106.2 and rotate everything. (source)

CVEs: MAL-2026-2144

[critical] Next.js out-of-band release fixes two unauthenticated RCEs (SVG image optimization CVE-2026-96889, use cache)

Fri, 23 Oct 2026 11:12:55 GMT

On Oct 14 2026 Vercel shipped Next.js 16.4.1 and 15.5.28 with two critical unauthenticated remote code execution fixes held back from the September release: CVE-2026-96889 / GHSA-2464-fhr4-w9xw, a use-after-free in librsvg (via sharp) reachable through the Image Optimization API when SVGs are processed on glibc-based Linux, and GHSA-673q-mjm6-r5xr, RCE in App Router apps with Cache Components when a `use cache` function returns multiple typed arrays. A high-severity Server Actions DoS (GHSA-p3m5-59mp-jv5r) is fixed too. GitHub added the SVG advisory to its global database on Oct 22. No known exploitation.

Status: live · Patch: patched · Fixed in: next 15.5.28, next 16.4.1, sharp 0.35.5 (librsvg 2.63.2)

Latest updates:

2026-10-22: GitHub publishes GHSA-2464-fhr4-w9xw to the global Advisory Database (critical), so Dependabot and npm audit start flagging affected next versions. (source)
2026-10-14: Vercel releases Next.js 16.4.1 and 15.5.28 as an out-of-band security update: two critical unauthenticated RCEs (SVG Image Optimization, `use cache` typed arrays) and a high-severity Server Actions DoS. No workaround for the `use cache` bug; upgrade even if you don't use those features. (source)
2026-10-14: Netlify: the SVG flaw is also reachable through `next/og` ImageResponse; on Netlify the next/image path runs on its Image CDN, and impact there is a crashed function, not code execution. (source)

CVEs: CVE-2026-96889

[medium] Nikkei breach: stolen credentials expose 1.83M personal-information records from a development cloud

Fri, 23 Oct 2026 11:12:55 GMT

Japanese media group Nikkei Inc. (owner of the Financial Times) says attackers used stolen credentials, obtained by unknown means, to access a cloud service it uses for system development in the early hours to morning of October 7, 2026, and personal information stored there leaked. As of October 16, about 1.83 million items had leaked: about 150,000 contain a name, email address, postal address, phone number or date of birth (about 120,000 include an email address), and about 1.68 million contain only information that cannot identify a person. No credit card data. Nikkei disabled the credentials, reported to Japan's Personal Information Protection Commission on October 11 and says it found no misuse so far. On October 21 Nikkei BP said about 800,000 of the leaked items belong to users of its services. BleepingComputer counts this as Nikkei's third disclosed incident in about a month.

Status: live · Patch: patched · Fixed in: Nikkei disabled the stolen credentials used for the access; scope review ongoing

Latest updates:

2026-10-23: Nikkei updates its breach notice: it modified its inquiry/contact form in response to a vulnerability identified in the PHP programming language. (source)
2026-10-21: Nikkei BP says about 800,000 of the leaked items belong to users of its services (about 50,000 with name, email, address, phone or date of birth; about 750,000 identifiers only). (source)
2026-10-16: Nikkei discloses the breach: about 1.83 million items leaked, about 150,000 with identifying details, no credit card data, no misuse or system outage found. It will contact affected people individually. (source)

[critical] Node2Shell: Node.js Brotli RCE (CVE-2026-111497) + October security releases

Fri, 23 Oct 2026 11:12:55 GMT

Node2Shell (CVE-2026-111497, also written Node 2 Shell) is a Brotli decoder out-of-bounds write in Node.js that can let an attacker run code when an app decompresses attacker-controlled Brotli data: Express 5 / body-parser 2.x request bodies sent with Content-Encoding: br, Brotli responses read by fetch(), or node:zlib on untrusted input. OpenAI Codex Security (Khai Tran and Sylvie Mayer) reported it; Node.js fixed it on Oct 21 2026 in 26.12.1, 24.22.1 and 22.23.4, alongside nine other CVEs (TLS hostname/IP bypasses, DNS NUL truncation) and a CA store refresh for the eMudhra incident. On Oct 22 Imperva Red Team, which coined the name Node2Shell, ran code on the official Node.js 24.22.0 image. NVD lists CVSS 4.0 9.2 (critical); Node.js rates it High. End-of-life lines (Node.js 20, 23, 25 and older) get no fix: upgrade to 26.12.1, 24.22.1 or 22.23.4. Bun is fixed in 1.4.4.

Status: live · Patch: patched · Fixed in: 22.23.4, 24.22.1, 26.12.1, Bun 1.4.4

Latest updates:

2026-10-22: End-of-life Node.js 20, 23 and 25 get no fix. Anyone on them has to move to 26.12.1, 24.22.1 or 22.23.4. Bun shipped the fix in 1.4.4. (source)
2026-10-22: NVD record for CVE-2026-111497 updated: the Google CNA scores it CVSS 9.2 (critical) and places the root bug in Google Brotli 1.0.3 through 1.2.0 (c/dec/decode.c), so other software bundling Brotli needs updating too. (source)
2026-10-22: Imperva Red Team (Ron Masas) names the bug Node2Shell and runs code on the official Node.js 24.22.0 image (Linux arm64) through fetch() and Express 5 / body-parser 2.x. It took several attempts; 24.22.1 rejected the same input. (source)

CVEs: CVE-2026-111497, CVE-2026-105482, CVE-2026-105483, CVE-2026-106069, CVE-2026-105481, CVE-2026-106072, CVE-2026-106074, CVE-2026-106070, CVE-2026-106071, CVE-2026-106073

[high] GMO Onamae Desktop Cloud malware on customer VPS

Fri, 23 Oct 2026 11:12:55 GMT

GMO Internet’s Japanese VPS service Onamae Desktop Cloud said an attacker installed malicious programs in customer environments, forcing shutdown of as many as ~26,739 servers with no immediate restart date (mid-October 2026 reporting).

Status: live · Patch: partial

[high] OpenCode Console breach exposes users' saved OpenAI / Anthropic / Google API keys

Fri, 23 Oct 2026 11:12:55 GMT

On the night of Oct 16 2026 an intruder reached a legacy OpenCode Console system that was being decommissioned and may have exposed third-party AI provider keys saved in Console workspaces. OpenCode says 1.4% of Console users (those with external providers linked) were affected and tied it to a large-scale distillation attack. Users reported unauthorized charges up to ~$1,700 and some Google Cloud project suspensions. Anthropic and Google revoked/blocked exposed keys; OpenAI refunded drawn credits. The OpenCode app itself (local keys) was not affected.

Status: live · Patch: partial

Latest updates:

2026-10-20: Anthropic credit restoration still in progress; OpenCode says it will cover them itself if not done this week. (source)
2026-10-19: BleepingComputer and The Hacker News report unauthorized charges up to ~$1,700 and Google Cloud project suspensions; OpenAI says it refunded affected credits. (source)
2026-10-18: Dax says attackers have been 'non stop' attacking OpenCode's systems for 48 hours; systems being hardened. (source)

[critical] OpenCode: malicious MCP server can run code on Windows/WSL during sign-in (no CVE)

Fri, 23 Oct 2026 11:12:55 GMT

A command-injection flaw in the OpenCode AI coding agent let a malicious or later-hijacked remote MCP server run PowerShell commands on a developer's Windows or WSL machine when they started the MCP OAuth sign-in. The server-supplied authorization endpoint was passed unescaped to an old version of the open package. Anomaly fixed it in 1.18.33 on Sep 28 without a CVE or advisory, so scanners won't flag 1.0.137 through 1.18.32. macOS and Linux are not affected. Pillar Security found it and rates it 9.8; no in-the-wild exploitation reported.

Status: live · Patch: patched · Fixed in: OpenCode 1.18.33

Latest updates:

2026-10-22: The Hacker News reports the flaw; upgrade to 1.18.33, run `opencode mcp list`, and remove unvetted remote MCP servers. (source)
2026-10-19: Pillar Security discloses the flaw publicly. Anomaly declined to request a CVE, saying user-configured MCP servers are outside its trust boundary. (source)
2026-09-28: OpenCode 1.18.33 ships the fix as a routine browser-launch refactor; release notes don't mention security. (source)

[high] Palo Alto Networks PAN-OS: Palo Alto Networks PAN-OS Code Injection Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Palo Alto Networks PAN-OS contains a code injection vulnerability that could allow an authenticated attacker with network access to execute arbitrary code with root privileges on the firewall. Added to CISA KEV on 2026-10-14.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-14: CISA added CVE-2026-0322 to the KEV catalog (due 2026-10-17).

CVEs: CVE-2026-0322

[critical] PHP2Shell: critical unauthenticated RCE in PHP core

Fri, 23 Oct 2026 11:12:55 GMT

Two critical remote-code-execution flaws in PHP core (CVE-2026-109618 and CVE-2026-109619, aka “PHP2Shell”) were disclosed early after an embargo leak. Both score CVSS 9.2 and can let an unauthenticated attacker run code on servers serving PHP web requests, especially Apache mod_php. Patches shipped Oct 21 2026 (appliance vendors such as QNAP are still preparing fixes); exploitation attempts were observed before and after the patch, and watchTowr / AWS report active scanning.

Status: live · Patch: partial · Fixed in: 8.5.12, 8.4.27, 8.3.36, 8.2.35

Latest updates:

2026-10-23: The PHP Foundation advises keeping session.upload_progress.enabled=0 until the next PHP release to block a further, unlikely RCE vector (CVE-2026-115560). It already recommended this setting for CVE-2026-109619; only apps with a non-default serialize_handler that switch it to "php" at runtime can be affected. (source)
2026-10-23: Palo Alto Networks published PAN-OS hotfix versions with ETAs from Oct 26 to Oct 30 depending on the branch (for example 12.2.4-h1 on Oct 28, 11.2.10-h16 and 10.2.18-h13 on Oct 26). Panorama and WildFire WF-500 appliances are still affected with no ETA. Until then, enable Threat IDs 543004 and 97622-97625 and lock down management access. (source)
2026-10-23: Sophos and Arctic Wolf see PHP2Shell exploit attempts trying to install Potassium, a Mirai-derived DDoS botnet; SOCRadar and Hunt.io found attacker servers with exploit tooling for PHP 8.4.26. None of these attempts got a shell, and Previdian, Wordfence, Beelzebub and Dataminr also report no successful RCE so far. Censys still counts over 34.4M vulnerable PHP services as of Oct 23. (source)

CVEs: CVE-2026-109618, CVE-2026-109619

[high] PingCAP / TiDB Cloud breach: stolen service-account credentials, data and source code taken

Fri, 23 Oct 2026 11:12:55 GMT

PingCAP confirmed that between Sep 9 and Oct 7 2026 an unauthorized party used stolen service-account credentials to access internal systems, copy data and obtain source code. Exported records included TiDB Cloud user names/emails, org/project associations, SQL-editor text, billing/contract data and an encrypted API-key table. Extortion group FulcrumSec claims 10.6 TB and 115k plaintext API keys; those figures are unverified and PingCAP disputes the API-key claim. No evidence data stored in TiDB was affected.

Status: live · Patch: partial

Latest updates:

2026-10-21: PingCAP Japan adds support-ticket information to the list of exposed data. (source)
2026-10-20: FulcrumSec goes public with claims (10.6 TB, 180k orgs, 115k API keys) and a ~200 GB sample; Bitget says it held only public trade logs there. (source)
2026-10-19: PingCAP publishes incident report; Japan notice cites up to ~50,000 personal records. (source)

[critical] Plex2Shell: chained Plex Media Server RCE exploited for miners

Fri, 23 Oct 2026 11:12:55 GMT

Four Plex Media Server flaws (CVE-2026-96652/96654/96655/96656), together called Plex2Shell, chain to unauthenticated admin-token theft and code execution. Researcher Zach Main published details Oct 16 and a PoC shortly after; exploitation of internet-facing servers (cryptominers + residential proxies) began by Oct 17. Fixed in 1.43.3.10861 (released earlier in August); CISA added the chain to KEV with a federal deadline of Oct 23.

Status: live · Patch: patched · Fixed in: 1.43.3.10861+

Latest updates:

2026-10-21: CISA KEV warning; patch-by Oct 23 for federal agencies.
2026-10-20: CISA adds CVE-2026-96652, -96654, -96655, -96656 to KEV (federal due dates Oct 23 for -96654/-96656, Nov 3 for -96652/-96655). (source)
2026-10-17: Honeypot exploitation (XMRig + proxy) observed.

CVEs: CVE-2026-96652, CVE-2026-96654, CVE-2026-96655, CVE-2026-96656

[critical] PwnEscape: Linux networking guest-to-host escape (CVE-2026-99999)

Fri, 23 Oct 2026 11:12:55 GMT

Paulos Yibelo / pwn.ai found a Critical Linux networking flaw (CVE-2026-99999, CVSS 9.3) during Vercel’s $1M Sandbox challenge. Guest-controlled checksum metadata can rewrite an IPv4 header after validation, enabling host-memory disclosure and, in lab conditions, guest-to-host code execution on KVM/Firecracker. Upstream fix merged Oct 2026; host kernels need patch + reboot. Affects multi-tenant VPS / sandbox providers more than end-user desktops.

Status: live · Patch: patched · Fixed in: Upstream commit 6785011f8b16 and matching stable backports; vendor kernels as listed by distros

Latest updates:

2026-10-14/15: Mainline merge / public write-ups.
2026-10-07/08: Upstream IPv4 checksum guard accepted into net.git.
2026-08-20: Initial host-memory disclosure reported.

CVEs: CVE-2026-99999

[high] Quest Diagnostics breach: patient records accessed via Quanum Lab Services Manager

Fri, 23 Oct 2026 11:12:55 GMT

Quest Diagnostics says it learned on Oct 21 2026 of unauthorized access to its Quanum Lab Services Manager (LSM), the system some clients use to place lab test orders and receive and look up results. Quest believes some patient data in Quanum LSM was accessed and confirms PHI/PII was involved; the specific data affected is still being investigated. Quest paused access to Quanum LSM (and, as a precaution, Essential Insights and LabHQ, where no data was accessed), notified federal law enforcement and engaged outside experts. Core operations are not materially disrupted, and Quest says the incident is not related to the PHP vulnerabilities.

Status: live · Patch: partial · Fixed in: Access to Quanum LSM paused and the unauthorized access discontinued; full restoration expected as soon as Oct 23

Latest updates:

2026-10-22: 8-K filing: Quest believes an unauthorized party accessed certain patient records; federal law enforcement notified and outside experts engaged; operations not materially disrupted. (source)
2026-10-22: Quest FAQ confirms PHI/PII was involved, pauses Essential Insights and LabHQ as a precaution, says the incident is not PHP-related, and expects Quanum LSM back as soon as Oct 23. Clients can use paper requisitions or Quanum eLabs meanwhile. (source)
2026-10-21: Quest learns of unauthorized access to Quanum LSM, starts incident response and pauses access to the system. (source)

[high] Railway discloses unauthorized access to customer environment variables

Fri, 23 Oct 2026 11:12:55 GMT

On Oct 21 2026 Railway reported anomalous activity on systems storing customer environment variables. An unauthorized third party may have accessed encrypted env vars and, for some, obtained decrypted contents (API keys, DB/registry credentials). Railway says it believes this is part of a series of attacks on platform companies aimed at compromising AI inference providers, and asked upstream providers to revoke keys. Builds were paused platform-wide for ~a day; on Oct 22 they were restored for Enterprise and Pro, with Hobby/Free/Trial still pending. All users are told to rotate secrets while scoping continues.

Status: live · Patch: partial

Latest updates:

2026-10-22: Builds restored for Enterprise (by ~06:12 UTC) and Pro (06:29 UTC); Hobby, then Free/Trial to follow. Deploys still queued/slow as of 08:44 UTC. (source)
2026-10-22: Railway publishes per-service remediation guides (AWS, GCP, GitHub, Stripe, OpenAI, Anthropic, Grok, OpenRouter, Postgres/MySQL/Redis/MongoDB credential rotation). (source)
2026-10-22: CEO: builds ramping up overnight; full post-mortem promised. (source)

[critical] Ruby 4.0.9, 3.4.12 and 3.3.13 fix Regexp compiler memory bugs behind GitLab's CVSS 10 RCEs

Fri, 23 Oct 2026 11:12:55 GMT

Ruby released 4.0.9, 3.4.12 and 3.3.13 on Oct 23 2026 to fix bugs in its Onigmo regular-expression engine (a double free and integer overflows) that GitLab turned into remote code execution on its servers, tracked by GitLab as CVE-2026-89078, CVE-2026-93577, CVE-2026-97229 and CVE-2026-110264. Any Ruby app that compiles a regex from an attacker-controlled string (Regexp.new, Regexp.union, Regexp.compile or /#{input}/) is exposed; Regexp.timeout and pattern-length limits do not protect. Apps that only use their own hard-coded regexes are not affected. Ruby 3.2 and older are end-of-life and get no fix.

Status: live · Patch: patched · Fixed in: 4.0.9, 3.4.12, 3.3.13

Latest updates:

2026-10-23: Ruby releases 4.0.9, 3.4.12 and 3.3.13 with the upstream fixes: a double free when parsing fails after an alternation (#13332), an integer overflow in the compiled pattern size (#19020) and an overflow of internal IDs in patterns with many repeats (#16801). Ruby requested no CVEs of its own and says to track GitLab's IDs in scanners. Credit: Johan Carlsson (joaxcar) via GitLab's bug bounty; GitLab reported and coordinated. (source)
2026-10-23: Ruby 3.3 is in security maintenance and gets the Regexp fixes as an exception. 3.3.13 also updates the default resolv gem to 0.3.2, fixing CVE-2026-80212 and CVE-2026-80213. (source)
2026-10-21: GitLab 19.5.1 and backports fix CVE-2026-110264 (CVSS 10.0, unauthenticated RCE from an integer overflow when compiling a crafted regex), affecting every CE/EE version from 7.0. (source)

CVEs: CVE-2026-89078, CVE-2026-93577, CVE-2026-97229, CVE-2026-110264

[high] Siemens Software Center: Siemens Software Center Uncontrolled Search Path Element Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Siemens Software Center contains an uncontrolled search path element vulnerability that could allow an authenticated local attacker to place a specially crafted DLL in a directory that the application searches, leading to arbitrary code execution with the privileges of the affected application. Added to CISA KEV on 2026-10-20.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-20: CISA added CVE-2026-110936 to the KEV catalog (due 2026-11-03).

CVEs: CVE-2026-110936

[medium] State Bank of India confirms customer data exposed on the dark web

Fri, 23 Oct 2026 11:12:55 GMT

State Bank of India (SBI), India's largest bank, disclosed that customer data turned up on the dark web. In a preliminary analysis of about 375,000 records, it identified 3,897 customer records with personal data: registered mobile numbers, dates of birth, names, gender, email addresses, residential addresses, branch and CIF identifiers, lead IDs and loan/lead metadata. SBI says balances and funds are safe and core operations are unaffected; the source of the exposure is still under investigation. Affected customers have been notified by email and/or WhatsApp. Expect targeted phishing and fraud calls.

Status: live · Patch: partial · Fixed in: SBI activated incident response and containment and engaged a CERT-In-empanelled agency; source and full extent still under investigation

Latest updates:

2026-10-23: SBI says it has individually informed all affected customers by email and/or WhatsApp and reiterates that account balances and funds are safe. (source)
2026-10-21: SBI discloses the incident to stock exchanges: 3,897 customer records with PII identified among about 375,000 records, containment activated and an independent CERT-In-empanelled agency engaged. (source)
2026-10-16: SBI receives alerts from its threat intelligence provider, the Defence Cyber Agency and CERT-In about suspected exposure of customer data on the dark web. (source)

[critical] tensorlake@0.5.144 npm Shai-Hulud worm with hostage GitHub token

Fri, 23 Oct 2026 11:12:55 GMT

On Oct 8 2026, tensorlake@0.5.144 was published from the real GitHub repo (with npm provenance) carrying a preinstall hook that installs a Shai-Hulud-style credential-stealing worm. It harvests npm/GitHub/cloud/SSH/AI-tool secrets, spreads to other packages, and installs a “gh-token-monitor” that wipes the home directory if the stolen GitHub token is revoked. Pin 0.5.143; remove the monitor before rotating tokens.

Status: live · Patch: partial · Fixed in: 0.5.143 (safe pin); remove 0.5.144

Latest updates:

2026-10-08: 0.5.144 published via the project’s release workflow.
2026-10-08+: StepSecurity / Socket detections and public write-ups; remediation guidance emphasizes removing gh-token-monitor before revoke.
2026-10-07: Malicious commits land on tensorlakeai/tensorlake main.

CVEs: GHSA-8g63-53c9-65j2

[medium] Major U.S. banks detect unauthorized third-party AI agents inside customer accounts

Fri, 23 Oct 2026 11:12:55 GMT

Axios reports that major U.S. banks, including Morgan Stanley, Bank of America, JPMorgan and Wells Fargo, are detecting third-party AI agents operating inside customer accounts. In most cases the agent logs in with the customer's own credentials and permission, which breaks bank rules that require people to log in themselves. Some banks also saw agents contact call centers or take actions that appeared independent of user instructions, and executives say they often can't tell a human from an agent, or an authorized agent from a rogue one. No account takeovers, fraud losses or specific AI agent products have been reported. Morgan Stanley is blocking third-party agents and JPMorgan, Bank of America and Citi are restricting them. Don't give an AI agent your bank username and password.

Status: live · Patch: partial · Fixed in: Morgan Stanley: blocking third-party agents, JPMorgan, Bank of America, Citi: restricting agent access, Most banks: moving to read-only access through authorized data connectors

Latest updates:

2026-10-22: Axios scoop: major U.S. banks are detecting unauthorized third-party AI agents inside customer accounts, some taking actions that appear independent of user instructions. Morgan Stanley is blocking agents and JPMorgan, Bank of America and Citi are restricting them. (source)
2026-10-22: Chase tells Axios that customers should not give their usernames and passwords to anyone. Morgan Stanley says it hasn't observed "rogue AI agents accessing our platforms." (source)
2026-10-22: Reuters picks up the Axios report and says it could not immediately verify it. (source)

[unknown] Refrigeration outages at up to a dozen US military commissaries, investigated by Army CID's Cyber Field Office (cause unconfirmed)

Fri, 23 Oct 2026 11:12:55 GMT

Starting Aug 26, refrigeration at Defense Commissary Agency (DeCA) stores on at least six US bases failed within days of each other, with some freezers going into defrost mode overnight and thousands of dollars of meat, dairy, produce and frozen food thrown out. Army CID's Cyber Field Office and the Air Force Office of Special Investigations confirmed investigations, with the Defense Criminal Investigative Service in the lead. The Pentagon only acknowledges a "possible refrigeration disruption" and no cause has been announced, so this is not a confirmed cyberattack; Reuters later noted it resembles the October attacks on French supermarket refrigeration controllers.

Status: live · Patch: unpatched

Latest updates:

2026-10-13: After the French supermarket refrigeration attacks, Reuters notes they resemble the late-August disruptions at about a dozen US military commissaries. (source)
2026-09-30: DeCA says cooler outages at four commissaries in Germany and England are equipment failures unrelated to the stateside outages; no findings from the investigation have been announced. (source)
2026-09-01: Army CID's Cyber Field Office and Air Force OSI confirm active investigations, with the Defense Criminal Investigative Service in the lead; some stores reported their freezers went into defrost mode. (source)

[critical] x402 SDK flaws could let malicious APIs drain AI agents' token balances

Fri, 23 Oct 2026 11:12:55 GMT

V12 disclosed 25 flaws in the x402 agent-payment SDKs, three rated critical: the default client had no spending limit (a hostile API could ask for an agent's whole token balance), the TypeScript Solana facilitator accepted token approvals as payments, and the SIWx login helper signed messages for other websites. All three were fixed in the August 18 releases, but no advisory was published, so npm audit and PyPI don't warn. Update @x402/fetch, @x402/axios, @x402/extensions and @x402/svm to 2.23.0 or later.

Status: live · Patch: patched · Fixed in: @x402/fetch, @x402/axios, @x402/extensions, @x402/svm 2.23.0+ (latest 2.28.0), x402 (PyPI) 2.25.0 (latest), Go module v2.28.0 (latest)

Latest updates:

2026-10-23: The Hacker News found the foundation published no security advisory for the three flaws, so npm audit and PyPI report nothing for the last vulnerable releases. Patched clients now cap each payment at $1 and allow only recognized USD stablecoins unless the app sets its own spend controls. (source)
2026-10-22: V12 published 25 x402 findings with three critical ones detailed: no default spend limit, Solana ApproveChecked accepted as TransferChecked in the TypeScript facilitator (a complete payment bypass in local tests), and SIWx signing relayed login messages for other domains. The x402 Foundation had fixed all three in the August 18 releases. (source)

[high] Zammad GmbH Zammad: Zammad GmbH Zammad Improper Privilege Management Vulnerability

Fri, 23 Oct 2026 11:12:55 GMT

Zammad GmbH Zammad contains an improper privilege management vulnerability that can allow the local zammad user to escalate privileges to root. This vulnerability can be chained with CVE-2026-102489. CISA added CVE-2026-102490, CVE-2026-102489 to the Known Exploited Vulnerabilities catalog on 2026-10-02 based on evidence of active exploitation.

Status: live · Patch: patched · Fixed in: See vendor advisory linked in sources

Latest updates:

2026-10-02: CISA added CVE-2026-102490, CVE-2026-102489 to the KEV catalog (due 2026-10-05).

CVEs: CVE-2026-102490, CVE-2026-102489

[critical] Zeabur breach exposes 3.2M env-var records; company to shut down

Fri, 23 Oct 2026 11:12:55 GMT

Zeabur disclosed an October 14–15 2026 security incident in which an attacker used leaked internal credentials stored in plaintext to access shared clusters in 9 regions, root into 101 Zeabur-managed GCP dedicated servers, and read the platform database, exposing ~3.22 million environment-variable records across 321,810 accounts. Zeabur is winding down: from December 1 2026 shared-cluster services pause and dedicated servers stop renewing, and on December 31 all services stop and service data is deleted. Refund and incident-claim requests close November 30. Users must rotate every secret stored on Zeabur and migrate off.

Status: live · Patch: partial

Latest updates:

2026-10-15: Attacker reads ~53 production secrets and re-adds SSH keys after a key is disabled; last recorded activity at 21:05. (source)
2026-10-14: Attacker starts using Zeabur internal AWS keys and shared-cluster admin tokens (stored in plaintext), reads secrets and env vars across 9 regions, and signs in as root to 101 GCP dedicated servers. (source)

[critical] Signal calls can stay live after hang-up (RingRTC CVE-2026-113380)

Thu, 22 Oct 2026 16:00:43 GMT

A bug in RingRTC, the open-source library behind Signal's voice and video calls, lets the other party on a call flood the call's state-machine event queue so a local hang-up gets stuck behind the backlog, and audio (possibly video) keeps flowing to them after the victim thinks the call ended. Affects every Signal client (Android, iOS, Desktop) embedding RingRTC before 2.72.0. Fixed in RingRTC 2.72.0 (released Sep 4 2026, disclosed Oct 20); a conceptual PoC is public, no known exploitation.

Status: closed · Patch: patched · Fixed in: RingRTC 2.72.0 (update Signal on every device)

Latest updates:

2026-10-22: No known exploitation; CISA has not added it to KEV and NVD lists it as awaiting analysis. (source)
2026-10-22: SecurityWeek: the hang-up was queued behind peer events with no rate limit or priority; the fix makes call termination bypass the queue. Users should update Signal and check the call indicator after hanging up. (source)
2026-10-22: Closed on pwn.watch: the fix shipped Sep 4 (RingRTC 2.72.0), there is still no known exploitation and no new developments, so this is now a patch-and-move-on item. Make sure Signal is updated on every device. (source)

CVEs: CVE-2026-113380

[high] Storybook dev server: malicious websites can inject code into your project (CVE-2026-114125)

Thu, 22 Oct 2026 12:43:45 GMT

Storybook disclosed CVE-2026-114125 on Oct 21 2026: the dev server's preview iframe could be embedded by any website and accepted postMessage events from any origin. A malicious site open in the same browser could, with some user interaction, make a running `storybook dev` server change story files in your project, and a changed file could then be executed by a local runtime with your privileges. Storybook 8, 9, 10 and 11 alpha are affected (including @storybook/addon-mcp and @storybook/addon-vitest). Fixed versions shipped Oct 12; Storybook says it is not aware of exploitation. Built (static) Storybooks are not affected.

Status: closed · Patch: patched · Fixed in: 8.6.19, 9.1.21, 10.6.2, 11.0.0-alpha.6

Latest updates:

2026-10-22: Closed on pwn.watch: fixes are out for every affected major and there is no known exploitation. Upgrade Storybook (and the Visual Tests addon to 5.5.1+ if you use it) and don't run an old dev server meanwhile. (source)
2026-10-21: Advisory GHSA-f2ph-cmg5-wxc8 / CVE-2026-114125 published. Storybook says it is not aware of exploitation in the wild; upgrade before running storybook dev again. (source)
2026-10-12: Patched releases 8.6.19, 9.1.21, 10.6.2 and 11.0.0-alpha.6 ship. They block cross-host iframe embedding of the dev server, reject postMessage from unknown origins and rework the highlight event (new feature flags can re-enable the old behaviour). (source)

CVEs: CVE-2026-114125

[critical] SubQuery @subql/common 5.8.3 published with credential stealer + backdoor

Thu, 22 Oct 2026 10:04:00 GMT

On Oct 5 2026 @subql/common 5.8.3 was published to npm through SubQuery's own GitHub Actions trusted publisher, after a short-lived branch changed the release step to swap the built package for a tarball downloaded from ci-artifacts.dev. The payload ran on install (postinstall) and on import, stole CI/cloud/GitHub/npm/Kubernetes/Vault/SSH and AI-tool credentials, offered a reverse shell, and with a workflow-scoped token tried to push a secret-dumping workflow. It was live about 50 minutes; 5.8.2 is the safe version. Many @subql packages accept 5.8.3 through their version ranges.

Status: closed · Patch: patched · Fixed in: @subql/common 5.8.2 (npm 'latest')

Latest updates:

2026-10-10: SubQuery maintainer: compromised token deleted, security review started. (source)
2026-10-06: GMO Flatt Security publishes payload analysis and response guidance. (source)
2026-10-05T11:56Z: Malicious @subql/common 5.8.3 published via the repo's GitHub Actions trusted publisher (canary 5.8.3-onf-rt1 published ~32 min earlier). (source)

CVEs: GHSA-9333-3c4x-x3h5, MAL-2026-17571

[medium] nginx 1.31.7 / 1.30.6 security fixes (HTTP/3, TLS tickets, overflows)

Thu, 22 Oct 2026 09:20:00 GMT

nginx published multiple medium/major advisories fixed in 1.31.7+ and 1.30.6+, including HTTP/3 request-body injection (CVE-2026-95101), TLS session-ticket forgery (CVE-2026-97629), AIO use-after-free (CVE-2026-97205), and buffer overflows in alias/rewrite and map/regex paths. Same-day chatter bundled this with the PHP/Node/GitLab rush for anyone terminating HTTP on nginx.

Status: closed · Patch: patched · Fixed in: 1.31.7+, 1.30.6+

Latest updates:

2026-10-21: Security advisories and patched releases published on nginx.org.
2026-10-21: nginx 1.31.8 mainline released. (source)
2026-10-14: nginx 1.30.6 stable and 1.31.7 mainline released with fixes for CVE-2026-15176, -93479, -95101, -97629, -97205; njs 1.0.2 fixes CVE-2026-101025. (source)

CVEs: CVE-2026-15176, CVE-2026-93479, CVE-2026-95101, CVE-2026-97629, CVE-2026-97205