The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has officially added two high-severity vulnerabilities to its Known Exploited Vulnerabilities (KEV) catalog, citing evidence that both are being actively leveraged in malicious campaigns. Federal Civilian Executive Branch agencies are required to remediate these flaws by November 16, 2026, in accordance with Binding Operational Directive 22-01.
The first vulnerability, tracked as CVE-2026-114713, affects UnrealIRCd, a widely used open-source IRC server software. Carrying a CVSS score of 8.6, this command injection flaw resides in the software’s labeled-response module. It impacts versions 6.0.0 through 6.2.7.0. The issue stems from a failure to properly sanitize IRCv3 label message tags before including them in server replies.
According to the software maintainers, this vulnerability allows an attacker to inject newlines and unauthorized commands into server-to-server communications. On networks utilizing at least two linked servers, a user without administrative privileges can exploit this to execute commands with server-level authority. This access could enable an attacker to perform privileged operations, such as banning users or disconnecting them from the network. The maintainers confirmed that exploitation attempts were observed on the internet shortly after the vulnerability was disclosed.
The second addition to the KEV catalog is CVE-2026-123039, a local privilege escalation vulnerability with a CVSS score of 7.8 affecting Check Point Security Gateways and Security Management products. The flaw is located within the Gaia OS DynamicCLI daemon, which operates with root privileges. The daemon exposes a world-writable Unix socket at /tmp/dynamic_socket without implementing proper authentication for connecting processes.
Check Point has confirmed that this vulnerability is being actively exploited in the wild. Attackers typically gain initial access to a target system by first exploiting separate PHP2Shell vulnerabilities, specifically CVE-2026-109618 and CVE-2026-109619, via the Mobile Access portal. Once inside, they utilize the DynamicCLI flaw to escalate their privileges to root.
Following successful privilege escalation, attackers have been observed deploying webshells and a backdoor known as TSGA to establish long-term persistence. This access allows for lateral movement within internal networks. In some instances, the attackers have deployed a wiper component, identified as destroy_v6, to delete data on VMware ESXi hosts.
The campaign involving the Check Point vulnerabilities appears to have unfolded in two distinct phases. The first wave of attacks was detected on October 30, 2026, primarily targeting organizations in Israel, with some incidents resulting in destructive data loss. A second, broader wave of attacks began on November 2, 2026, impacting organizations across multiple sectors and geographic regions.
While the tactics, techniques, and procedures observed in these attacks share code and infrastructure characteristics with the PhiliKit framework—previously linked to the threat actor group known as UNC5221—there is no definitive consensus on the identity of the attackers. Researchers have noted that the evidence is currently insufficient to confirm the involvement of any specific group, suggesting the possibility of a false-flag operation or collaboration between different actors.
As of now, no specific threat actor has been publicly identified for the UnrealIRCd exploitation campaign, and there is no confirmed evidence linking the activity to a particular financial or political motive. CISA and security experts strongly advise both federal agencies and private sector organizations to prioritize patching these systems to mitigate the risk of further unauthorized access and data destruction.





