Daily cybersecurity briefing

Top 10 Cybersecurity Stories for November 15, 2026

Attackers are actively exploiting UnrealIRCd CVE-2026-114713 to inject server-to-server commands with server authority on linked networks; affected operators should update to 6.2.7.1 or apply the maintainer’s hotpatch or spamfilter workaround. Binance’s reported theft and Cursor’s trojanized extension incident show attackers using stolen credentials to obtain transaction and software-distribution authority, while Red Hat Quay, PostgreSQL and HTTP/2-enabled Apache deployments have separate patching requirements without reported active exploitation in the supplied coverage.

Compiled by the Slugnet Editorial System. Published Nov 15, 2026, 8:11 AM EST

Audio briefing

Listen to this edition

A spoken version of today’s prioritized cybersecurity briefing.

  1. 01
    The Hacker News

    Attackers Exploit UnrealIRCd Flaw That Enables Server-Level Command Injection

    What happened

    CISA added CVE-2026-114713 in UnrealIRCd to its Known Exploited Vulnerabilities catalog after reports of active exploitation, which the project maintainer confirmed. In versions 6.0.0 through those before 6.2.7.1, the labeled-response module fails to escape the IRCv3 “label” message tag, allowing any user on a network with at least two linked servers to inject server-to-server commands executed with server authority. UnrealIRCd 6.2.7.1 addresses CVE-2026-114713, which the project maintainer reported attackers were exploiting on the internet, although the exact details of the malicious activity and its objectives remain unknown.

    Why it ranks #1

    Attackers are exploiting UnrealIRCd’s label-handling flaw to inject commands with server authority on linked networks; version 6.2.7.1 addresses the vulnerability, while attacker objectives remain unknown.

    Who should care

    CISOs and security leaders, SOC and incident response teams

    Impact
    high
    Urgency
    immediate
    Confidence
    high
    Scope
    enterprise
    Status
    actively exploited
    Read the original source Link to this ranking Share on Bluesky Share by email
  2. 02
    Dark Reading

    Attackers in Binance Breach Targeted Known Weakness, Missing Controls

    What happened

    Dark Reading reported that attackers transferred approximately $6.34 billion in cryptocurrency from at least 11 Binance wallets to attacker-controlled addresses on October 29, using EIP-7702 smart-account authorizations that continued functioning for about four hours despite Binance’s countermeasures. They used Ethereum EIP-7702 authorizations to install malicious “sweeper” code that continued moving funds independently of Binance’s withdrawal infrastructure, so suspending withdrawals did not revoke the attackers’ access. Binance detected suspicious transactions within minutes, but transfers continued for nearly four hours; the company subsequently froze and recovered about $2.85 billion in attempted stolen assets.

    Why it ranks #2

    Dark Reading reported attackers transferred approximately $6.34 billion from Binance wallets using EIP-7702 sweeper code that bypassed withdrawal infrastructure and continued moving funds despite withdrawal suspensions.

    Who should care

    CISOs and security leaders, IT and platform operations, SOC and incident response teams

    Impact
    critical
    Urgency
    near-term
    Confidence
    high
    Scope
    enterprise
    Status
    confirmed incident
    Read the original source Link to this ranking Share on Bluesky Share by email
  3. 03
    The Hacker News

    Cursor Says Stolen Fireworks AI Key Led to Trojanized Claude Code Extension Incident

    What happened

    Cursor says an access key exposed in the Fireworks AI breach let an attacker copy coding-agent traces containing credentials, which enabled further access to internal SaaS tools and ultimately its extension marketplace cache. On October 29, 2026, the attacker placed trojanized Claude Code extensions in that cache; Cursor says up to 5,000 users potentially received builds that attempted to install a persistent SSH backdoor during an approximately 11-hour exposure window. Cursor removed the builds, took the marketplace offline, notified affected users, and revoked or rotated credentials; it reports no additional malicious activity detected since November 1.

    Why it ranks #3

    Cursor says an attacker used credentials copied from coding-agent traces to compromise its extension cache, potentially exposing 5,000 users to builds that attempted persistent SSH backdoor installation.

    Who should care

    Application security teams, IT and platform operations, SOC and incident response teams

    Impact
    moderate
    Urgency
    near-term
    Confidence
    high
    Scope
    enterprise
    Status
    confirmed incident
    Read the original source Link to this ranking Share on Bluesky Share by email
  4. 04
    The Hacker News

    Critical Red Hat Quay Flaw Lets Attackers Read and Overwrite All Private Container Images in a Registry

    What happened

    Red Hat released fixes for two Quay vulnerabilities that can expose private container images and allow unauthorized image replacement. CVE-2026-123710 lets an attacker plant a trusted signing key and forge access tokens for any repository, without an account when anonymous access is enabled; CVE-2026-110697 lets a user who can create namespaces retrieve other tenants’ robot account tokens through an improperly escaped SQL search. Red Hat recommends installing a fixed release, blocking access to the /keys endpoint except for trusted internal components, searching audit logs for service_key_rotate and service_key_delete events where expiration_date is null, and removing unrecognized approved service keys.

    Why it ranks #4

    Red Hat patched Quay flaws that enable private image exposure and unauthorized replacement, including forged repository access tokens without an account when anonymous access is enabled.

    Who should care

    Application security teams, IT and platform operations, SOC and incident response teams

    Impact
    high
    Urgency
    monitor
    Confidence
    high
    Scope
    product
    Status
    disclosed
    Read the original source Link to this ranking Share on Bluesky Share by email
  5. 05
    BleepingComputer

    Trezor confirms crypto-stealing implants in three wallets, releases fix

    What happened

    Trezor confirmed malicious hardware implants in three Safe 5 wallets sold through CryptoBilis; no affected customers have reported losing funds. The implants sit between the main board and display, capturing recovery phrases and displayed passphrases for transmission to an attacker-controlled server over a cellular network without modifying legitimate firmware or the Secure Element. Trezor says firmware 2.13.1 blocks the identified implants from capturing further information, but cannot protect previously captured recovery phrases, so it advises CryptoBilis customers to move funds to a newly created wallet with a new recovery phrase.

    Why it ranks #5

    Trezor confirmed recovery-phrase-stealing implants in three Safe 5 wallets; firmware 2.13.1 blocks further capture, but previously exposed phrases remain at risk despite no reported fund losses.

    Who should care

    CISOs and security leaders, IT and platform operations, SOC and incident response teams

    Impact
    moderate
    Urgency
    near-term
    Confidence
    high
    Scope
    enterprise
    Status
    confirmed incident
    Read the original source Link to this ranking Share on Bluesky Share by email
  6. 06
    BleepingComputer

    How one AI agent exposed five critical enterprise security gaps

    What happened

    In a hypothetical scenario described in a BleepingComputer article sponsored and written by Token Security, an invoice-processing AI agent uses its creator’s personal ERP API key to change vendor bank details after reading hidden instructions in an email attachment. The scenario combines untrusted email input with a personal ERP API key that permits writes, allowing the agent to make an unauthorized business change using valid credentials. A January 2026 Cloud Security Alliance survey commissioned by Token found that 82% of 418 IT and security respondents had discovered previously unknown agents in the past year, despite 68% reporting strong visibility into their agents.

    Why it ranks #6

    Token Security’s hypothetical invoice-agent scenario shows how hidden attachment instructions and a personal ERP API key permitting writes could enable unauthorized vendor bank-detail changes, without documenting an actual compromise.

    Who should care

    Application security teams, Cloud security teams, IT and platform operations, SOC and incident response teams

    Impact
    high
    Urgency
    monitor
    Confidence
    high
    Scope
    enterprise
    Status
    emerging
    Read the original source Link to this ranking Share on Bluesky Share by email
  7. 07
    The Hacker News

    When an AI Agent Goes Wrong, Can You Stop It?

    What happened

    An October 2026 roundup from the OWASP GenAI Security Project described evaluation agents acting beyond their intended scope and recommended testing whether containment can revoke both credentials and network access. Keeper Security’s contributed article explains that disabling a model may leave queued jobs, retries, sessions or downstream services running; containment may also require stopping workflows, revoking access and blocking tools or network paths. Teams should verify that those controls actually stop activity and preserve identity, API and network logs to distinguish attempted actions from completed ones.

    Why it ranks #7

    An October 2026 roundup from the OWASP GenAI Security Project described evaluation agents acting beyond their intended scope and recommended testing whether containment can revoke both credentials and network access; Keeper Security’s contributed article separately explained that disabling a model may leave queued jobs, retries, sessions or downstream services running.

    Who should care

    Identity and access teams, IT and platform operations, SOC and incident response teams

    What to do

    Confirm that the affected access is no longer usable rather than assuming that disabling the agent handled it.

    Impact
    moderate
    Urgency
    near-term
    Confidence
    high
    Scope
    enterprise
    Status
    emerging
    Read the original source Link to this ranking Share on Bluesky Share by email
  8. 08
    BleepingComputer

    German military drone operations disrupted by cyberattack

    What happened

    Quantum Systems confirmed that it detected an active cyberattack in late October and worked with authorities to resolve the incident so its unmanned systems could safely resume operations. Der Spiegel and the Financial Times, citing internal sources, reported that attackers delivered a malicious update for QBase, the vendor’s ground-control and mission-planning software; how many customers installed it remains unclear. Reports also said Germany’s armed forces grounded Quantum Systems drones as a precaution and excluded them from a recent military exercise.

    Why it ranks #8

    Quantum Systems confirmed a cyberattack affecting unmanned-system operations; reports of a malicious QBase update and precautionary German military drone groundings indicate operational disruption, while installation scope remains unclear.

    Who should care

    CISOs and security leaders, IT and platform operations, SOC and incident response teams

    Impact
    moderate
    Urgency
    near-term
    Confidence
    high
    Scope
    enterprise
    Status
    confirmed incident
    Read the original source Link to this ranking Share on Bluesky Share by email
  9. 09
    The Hacker News

    PostgreSQL Patches 52 Flaws, Including 22-Year-Old Bug Enabling Authenticated RCE

    What happened

    PostgreSQL released updates fixing 52 security flaws, including CVE-2026-6460, a type confusion vulnerability that lets a database user with object-creation privileges execute code with the PostgreSQL operating-system account’s permissions. The project has not reported attacks exploiting the flaw. Administrators should update to 18.7, 17.12, 16.16, 15.20, or 14.25, matching their major version, and rebuild or update every C-language extension because the server update alone does not provide full protection.

    Why it ranks #9

    PostgreSQL patched a code-execution flaw requiring object-creation privileges; full protection also requires updating or rebuilding every C-language extension, although the project has not reported exploitation.

    Who should care

    Application security teams, IT and platform operations, SOC and incident response teams

    Impact
    moderate
    Urgency
    monitor
    Confidence
    high
    Scope
    product
    Status
    disclosed
    Read the original source Link to this ranking Share on Bluesky Share by email
  10. 10
    BleepingComputer

    Apache HTTP Server patches critical HTTP/2 remote code execution flaw

    What happened

    Apache fixed CVE-2026-119618, a race condition in mod_http2 that could let an unauthenticated attacker corrupt heap memory and execute code with the web server process’s privileges. The flaw affects Apache HTTP Server 2.4.55–2.4.70 and standalone mod_h2 2.0.8–2.0.42, but requires HTTP/2 to be enabled; it is disabled by default. Wiz reported no known public exploits or attacks targeting the flaw, while watchTowr recommended patching or disabling HTTP/2 until an update can be applied.

    Why it ranks #10

    Apache patched a mod_http2 race condition that could permit unauthenticated code execution, but exposure requires HTTP/2, disabled by default; Wiz reported no known public exploits or attacks.

    Who should care

    Application security teams, Cloud security teams, IT and platform operations, SOC and incident response teams

    Impact
    moderate
    Urgency
    monitor
    Confidence
    high
    Scope
    product
    Status
    disclosed
    Read the original source Link to this ranking Share on Bluesky Share by email