Verified status as of November 15, 2026
Checked by defend.network against the NIST National Vulnerability Database (NVD) and CISA’s Known Exploited Vulnerabilities catalog.
- CVE-2026-114713 in UnrealIRCd is listed in CISA’s Known Exploited Vulnerabilities (KEV) catalog (added 2026-11-13; federal remediation due 2026-11-16). The NVD record carries a CNA/ADP score of 8.6 (HIGH) on CVSS v3.1.
- CVE-2026-111497 in Google Brotli is listed in CISA’s Known Exploited Vulnerabilities (KEV) catalog (added 2026-10-27; federal remediation due 2026-11-10). NVD rates it 9.8 (CRITICAL) on CVSS v3.1.
- CVE-2026-109618 in PHP is listed in CISA’s Known Exploited Vulnerabilities (KEV) catalog (added 2026-10-26; federal remediation due 2026-10-29). NVD rates it 9.8 (CRITICAL) on CVSS v3.1. NVD lists a vendor advisory.
- CVE-2026-109619 in PHP is listed in CISA’s Known Exploited Vulnerabilities (KEV) catalog (added 2026-10-26; federal remediation due 2026-10-29). NVD rates it 8.1 (HIGH) on CVSS v3.1. NVD lists a vendor advisory.
TL;DR
UnrealIRCd server flaw actively exploited; Cursor code editor breached, exposing up to 5,000 developers to malware; a November 2 attack on Japanese email service Arara exposed 77 million email addresses, and the City of Osaka reported that over 1 million of its recipients’ addresses leaked. Patch immediately, rotate credentials, and monitor for lateral movement.
Today at a glance grouped by CISA KEV, NVD and source-reported status as of 2026-11-15
Exploited in the wild
UnrealIRCd Server Command Injection Under Active Exploitation — 1 CVE in CISA KEV; earliest federal deadline 2026-11-16
No CVE from this briefing’s verified CVE table – status as the sources report it
Cursor Code Editor Breach Leads to Malware Distribution Via Marketplace
Attack on Japanese Email Service Arara Exposes 77 Million Addresses; City of Osaka Affected
GitHub wfx-loot Campaign Compromises Thousands of Repositories via Stolen Accounts
Operation Endgame Dismantles Gootloader, CastleRAT, StealC Infrastructure; 6.4M Logins Seized
Executive Summary
- UnrealIRCd server flaw (CVE-2026-114713) is actively exploited in the wild; CISA added it to KEV catalog. Patch to 6.2.7.1 or deploy spamfilter workaround immediately.
- Cursor (AI code editor) suffered breach via Fireworks AI credential theft; malicious Claude Code extension deployed to marketplace, exposing up to 5,000 developers to persistent SSH backdoors.
- A November 2 attack on Japanese email distribution service Arara exposed email addresses from nearly 400 customers including Sega; over 1 million addresses of people who received City of Osaka emails leaked, and attackers used the service to send malicious emails.
- GitHub wfx-loot campaign used 4,837 compromised accounts targeting 61,764 repositories; researchers confirmed the malware ran and stole secrets in at least 1,835 repositories.
- Operation Endgame seized 6.4M stolen logins; took down Gootloader, CastleRAT, and StealC infrastructure across 235 servers and 257 domains.
Top Threats Today
1. UnrealIRCd Server Command Injection Under Active Exploitation
Severity: HIGH Affected: Technology
UnrealIRCd versions 6.0.0 through 6.2.7.0 contain a command injection flaw (CVE-2026-114713, CVSS 8.6) that allows any user on a network with two or more linked IRC servers to inject server-to-server commands executed with server authority [1]. The labeled-response module fails to escape the IRCv3 'label' message tag when echoing replies, creating the injection vector [1]. Exploitation attempts are said to have begun a day after the November 7, 2026 hotfix release; the maintainer confirmed “bad guys exploiting this issue on the internet right now” [1]. CISA added CVE-2026-114713 to its Known Exploited Vulnerabilities catalog, and Federal Civilian Executive Branch agencies have until November 16, 2026 to remediate under BOD 26-04 [1]. The maintainer described it as the “most critical issue” found in UnrealIRCd over the past 15 years [1].
Sources:[1] The Hacker News
Verified data · NVD, CISA KEV, FIRST EPSS · as of 2026-11-15
CVE-2026-114713 UnrealIRCd CVSS 8.6 HIGH (CNA/ADP, v3.1) in CISA KEV since 2026-11-13; federal deadline 2026-11-16 EPSS 1% (higher than 44% of scored CVEs), scored 2026-11-14
Earlier coverage: November 14, 2026 briefing
Recommended Action
- Update to UnrealIRCd 6.2.7.1 immediately
- If patching cannot be deployed at once, deploy the spamfilter workaround referenced by the maintainer
- Monitor linked IRC servers for anomalous server-to-server command activity and review access logs for exploitation attempts since November 7
2. Cursor Code Editor Breach Leads to Malware Distribution Via Marketplace
Severity: HIGH Affected: Technology
Cursor, a Microsoft Visual Studio Code-based AI code editor, suffered a breach initiated through credential theft from Fireworks AI, a third-party AI infrastructure provider [1]. Attackers obtained an access key to a read-only storage bucket containing historical AI coding-agent traces from users without Privacy Mode enabled [1]. These traces included unredacted API credentials, which attackers leveraged to breach additional Cursor systems and obtain marketplace distribution credentials [1]. On October 29, 2026, malicious versions of Anthropic's Claude Code extension were published to Cursor's marketplace, exposing up to 5,000 developers to persistent SSH backdoors installed by the malware [1]. Fewer than 0.05% of users had coding-agent traces exposed, and fewer than 0.01% had API credentials compromised, according to Cursor [1]. The malicious versions, 2.1.312 and 2.1.313, were available for about 11 hours before Cursor removed them, and the attackers also accessed internal systems and stole customer data [1].
Sources:[1] BleepingComputer
Recommended Action
- If you installed or updated the Claude Code extension from Cursor's marketplace during the roughly 11-hour window that began on October 29, 2026 (malicious versions 2.1.312 and 2.1.313), follow Cursor's remediation guidance
- Scan systems for unauthorized SSH keys or connections; check ~/.ssh/authorized_keys and SSH server logs for suspicious activity
- Rotate API credentials and OAuth tokens issued before the breach discovery; monitor Cursor and any connected services for unauthorized access
3. Attack on Japanese Email Service Arara Exposes 77 Million Addresses; City of Osaka Affected
Severity: HIGH Affected: Government Technology
The City of Osaka, Japan's third-largest city, disclosed that over 1 million email addresses tied to its online administrative services platform were exposed after hackers accessed the Arara email distribution service used by TKC Corporation, the contractor that maintains the platform [1]. Anyone who received an email from the city's system between July and October 31, 2026 had their email address leaked [1]. Arara said the November 2 cyberattack exposed 77 million email addresses from nearly 400 customers, including TKC and gaming giant Sega [1]. Attackers used Arara's infrastructure to send millions of malicious emails impersonating credit card companies and financial institutions [1]. The city urged its 2.8 million residents to be wary of emails purporting to come from the city and not to open URLs or attachments in them [1]. Japan's national Computer Security Incident Response Team published research three weeks prior identifying attackers using AI to scan for known vulnerabilities before exploitation [1].
Sources:[1] The Record
Recommended Action
- Osaka residents and Arara customers: flag suspicious emails claiming to be from financial institutions or requesting urgent action as likely phishing attempts
- Organizations using Arara services should contact Arara and audit logs for unauthorized access or message delivery since November 2
- Monitor user accounts for compromised credentials and enable MFA where available to prevent account takeover
4. GitHub wfx-loot Campaign Compromises Thousands of Repositories via Stolen Accounts
Severity: HIGH Affected: Technology
Attackers used 4,837 compromised GitHub accounts in the wfx-loot campaign to open probe pull requests across 61,764 repositories beginning November 10, 2026 [1]. Where access allowed, the attackers followed with malicious commits carrying a credential-stealing shell script; researchers confirmed the malware ran and stole secrets in at least 1,835 repositories, and OX Security’s November 12 report counted malicious commits in 8,093 public repositories [1]. The payload targets GitHub secrets, workflow tokens, cloud credentials, SSH/GPG keys, environment files, Kubernetes configurations, and credentials for AI services including OpenAI, Anthropic, and Google Gemini [1]. Attackers reverted their changes and deleted workflow runs and logs, sometimes within a minute [1]. Fly.io disclosed that tokens issued to its GitHub App and left in publicly served files of some deployed applications were used by an unauthorized party against customer repositories, which explains how some of the credentials used in the campaign were obtained [1].
Sources:[1] HackRead
Recommended Action
- Audit GitHub repository secrets, personal access tokens, and deployment workflows; rotate all exposed credentials
- Verify no unauthorized changes were made to GitHub Actions workflows or shell scripts; review commit history for November 10 onward
- Check cloud provider and AI service credentials for unauthorized use; enable MFA and IP allowlisting on critical GitHub accounts
5. Operation Endgame Dismantles Gootloader, CastleRAT, StealC Infrastructure; 6.4M Logins Seized
Severity: HIGH Affected: Technology Finance
An international law enforcement operation took down infrastructure behind three major malware tools: Gootloader (a loader used to distribute ransomware), CastleRAT (a remote access trojan), and StealC (an infostealer) [1]. Police seized 235 servers, 257 domains, and approximately 6.4 million stolen login credentials from more than 40,000 victims [1]. Gootloader has infected at least 25,000 computers worldwide since 2018, according to the U.S. Justice Department; it spread via SEO-poisoned fake legal documents and was used in U.S. ransomware attacks on at least 44 victims that caused about $28.8 million in losses [1]. CastleRAT, known since 2025, allows remote control of infected machines; StealC exfiltrates passwords and login data [1]. StealC's operators had rebuilt infrastructure after a June 2026 takedown and redeployed [1]. The operation included removal of Gootloader from more than 100 servers and 681 hacked websites [1].
Sources:[1] The Hacker News
Recommended Action
- Check individual email addresses in the Dutch police service “Check your hack”, which so far holds data stolen by CastleRAT and checks single addresses, not whole domains; organizations can join the police project No More Leaks. If an address is found, follow the removal steps the police send, reset passwords and audit account activity
- Scan for Gootloader via SEO poisoning indicators: search results pointing to unusual sites hosting fake documents; block known compromised domains from the takedown
- Monitor endpoints for StealC and CastleRAT indicators; review web traffic for connections to seized servers and domains
Ongoing Coverage
- Check Point, VMware, GitHub, Apache vulnerabilities: Earlier coverage on 2026-11-14 continues. No new material developments today.
- Microsoft security updates and FAA ground stop: BleepingComputer reports November 2026 Windows security updates are causing printing failures for documents with complex formatting including images with zero height/width. Separately, the FAA issued a nationwide ground stop on November 13 following a communications disruption; officials found no evidence of a cyberattack, with FBI investigating a software update as a potential cause.
- North Korean FaceStealer campaign: The U.S. disrupted the campaign responsible for $400 million in cryptocurrency theft. See 2026-11-14 briefing for full details.
Today’s Action Checklist
- ☐ URGENT: If running UnrealIRCd 6.0.0–6.2.7.0, update to 6.2.7.1 or deploy spamfilter workaround immediately (U.S. federal civilian agencies must apply the fixes by November 16, 2026 under BOD 26-04)
- ☐ URGENT: If you installed or updated the Claude Code extension in Cursor during the roughly 11-hour window that began on October 29, 2026, follow Cursor's remediation guidance; scan for unauthorized SSH keys
- ☐ HIGH: Audit GitHub account history and repository commits for November 10+ changes; rotate exposed credentials (wfx-loot campaign)
- ☐ HIGH: Check individual email addresses in the Dutch police service “Check your hack” (it holds data stolen by CastleRAT so far and does not check whole domains); reset exposed credentials if found
- ☐ If using Arara email services: audit logs for unauthorized access; monitor users for credential compromise
- ☐ Review printing configurations after applying November 2026 Windows updates; revert if complex document printing fails
CISA KEV watch from the CISA Known Exploited Vulnerabilities catalog as of 2026-11-15 · catalog data, not AI-written
Earlier additions with a federal deadline through 2026-11-22 (4)
CVE-2026-114713 UnrealIRCd federal deadline 2026-11-16 (added 2026-11-13)
CVE-2026-123039 Check Point Multiple Products federal deadline 2026-11-16 (added 2026-11-13)
CVE-2024-27834 Apple Multiple Products federal deadline 2026-11-18 (added 2026-11-04)
CVE-2026-19868 SonicWall SonicOS SSLVPN federal deadline 2026-11-20 (added 2026-11-06)