ThreatBeaconXSubscribe
HighVulnerability · 9 min read

CISA Adds Two Actively Exploited Vulnerabilities to KEV Catalog: Check Point Zero-Day and UnrealIRCd Command Injection

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) added two actively exploited vulnerabilities to its Known Exploited Vulnerabilities (KEV) Catalog on November 13, 2026. The additions include a local privilege-escalation vulnerability affecting multiple Check Point products and a command-injection vulnerability affecting UnrealIRCd. The vulnerabilities are: CVE-2026-123039: Incorrect Permission Assignment for a Critical Resource in multiple Check Point products. CVE-2026-114713: Improper Neutralization of Delimiters in UnrealIRCd. Both vulnerabilities have a federal remediation deadline of November 16, 2026. The Check Point vulnerability allows an authenticated local user to execute commands with root privileges through an improperly protected DynamicCLI socket in Gaia OS. The UnrealIRCd vulnerability allows an ordinary user on a network with at least two linked IRC servers to inject server-to-server commands that execute with server authority. CISA's additions highlight the importance of prioritizing vulnerabilities with confirmed exploitation evidence, applying vendor-provided fixes and investigating systems for possible compromise rather than treating patch installation as the only response.

Written by Vedanabhatla Sai Swetha·Published Nov 14, 2026

Description

CISA maintains the Known Exploited Vulnerabilities Catalog as a resource for identifying vulnerabilities that have evidence of exploitation in real-world environments.

On November 13, 2026, the agency added two vulnerabilities affecting Check Point security infrastructure and UnrealIRCd.

The first vulnerability, CVE-2026-123039, affects Check Point products running Gaia OS. The DynamicCLI daemon runs with root privileges and listens on a world-writable local socket without properly authenticating connecting processes. An authenticated local user can exploit this condition to execute commands as root.

The second vulnerability, CVE-2026-114713, affects UnrealIRCd's labeled-response module. Improper handling of IRCv3 label message tags allows a user to inject server-to-server commands into responses. On networks with two or more linked servers, these commands can execute with server authority.

Both vulnerabilities were added to the KEV Catalog based on evidence of active exploitation. CISA instructed affected organizations to apply vendor mitigations and perform the required forensic triage.

Technical Description

1. CVE-2026-123039 — Check Point Multiple Products Incorrect Permission Assignment for Critical Resource

<text>Severity: High — CVSS 7.8</text>

Affected vendor: Check Point Software Technologies

Affected products: Check Point Firewall, Multi-Domain Security Management Server, Scalable Platforms, Security Gateway and Security Management Server.

Vulnerability classification:

  • CWE-732 — Incorrect Permission Assignment for Critical Resource

  • CWE-862 — Missing Authorization

The vulnerability exists in the DynamicCLI inter-process communication (IPC) interface used by Gaia OS.

The DynamicCLI daemon runs as root and listens on a local Unix socket:

/tmp/dynamic_socket

The socket is world-writable and does not properly authenticate connecting processes.

An authenticated local user can send commands to this socket, which are then executed by the root-level daemon.

This creates a local privilege-escalation path. An attacker who already has local access to a vulnerable system can potentially elevate privileges from an ordinary authenticated user to root.

Security impact

Successful exploitation can provide an attacker with root-level command execution on the affected appliance. This can enable unauthorized system modifications, persistence and further malicious activity.

The vulnerability is particularly concerning because Check Point reported observing active exploitation against a couple of dozen customers.

2. CVE-2026-114713 — UnrealIRCd Improper Neutralization of Delimiters

<text>Severity: High — CVSS 8.6</text>

Affected vendor: UnrealIRCd

Affected product: UnrealIRCd 6

Affected versions: UnrealIRCd 6.0.0 before 6.2.7.1.

Vulnerability classification:

  • CWE-140 — Improper Neutralization of Delimiters

The vulnerability affects the labeled-response module in UnrealIRCd.

The module fails to properly escape the IRCv3 label message tag when echoing it into replies.

An attacker can exploit this improper neutralization to inject server-to-server commands into responses.

On an IRC network containing two or more linked servers, the injected commands can be executed with server authority.

According to the UnrealIRCd project maintainers, an ordinary user could abuse the vulnerability to execute commands with server or IRC operator privileges.

This can enable unauthorized administrative actions, including commands such as KILL and GLINE, which can disrupt IRC network operations and affect connected users.

Security impact

Successful exploitation can compromise the integrity of IRC network operations by allowing an ordinary user to perform actions normally restricted to privileged server administrators.

The vulnerability is especially serious for networks running vulnerable versions with multiple linked servers.

Attack Overview

Vulnerability 1: Check Point DynamicCLI Privilege Escalation

The reported exploitation path involves an attacker obtaining an initial foothold and then abusing the vulnerable DynamicCLI interface.

Attack sequence

  1. Initial access: Attackers gain access to a vulnerable Check Point appliance.

  2. Local execution: An authenticated local user can interact with the DynamicCLI socket.

  3. Privilege escalation: The attacker sends commands through the world-writable socket.

  4. Root execution: The root-level daemon executes the supplied commands with elevated privileges.

  5. Post-exploitation: An attacker can potentially establish persistence, deploy additional tools or move further into the environment.

Additional reporting from Check Point's incident investigation describes a broader attack chain involving PHP2Shell vulnerabilities in Mobile Access portals, followed by CVE-2026-123039 exploitation.

The reported activity included webshell deployment, TSGA backdoor and TSProxy malware installation, movement to VMware ESXi infrastructure and deployment of destructive wiper malware.

These details describe the reported Check Point exploitation activity and should not be interpreted as proof that every vulnerable appliance experienced the entire attack chain.

Vulnerability 2: UnrealIRCd Server-to-Server Command Injection

The UnrealIRCd vulnerability abuses improper handling of the IRCv3 label message tag.

Attack sequence

  1. Access to an IRC network: An attacker connects as an ordinary user.

  2. Malicious message construction: The attacker crafts a message containing a specially formatted label value.

  3. Improper processing: The labeled-response module fails to escape the label correctly when generating a reply.

  4. Command injection: The crafted content is interpreted as server-to-server commands.

  5. Server-authority execution: On a network with at least two linked servers, the injected commands execute with server authority.

The UnrealIRCd maintainers publicly warned that exploitation attempts were being observed and that vulnerable networks should apply the workaround and update or hot-patch their servers.

Technical Analysis

Check Point: DynamicCLI Socket Abuse

The central issue is the mismatch between the privileges of the DynamicCLI daemon and the permissions of the local socket it exposes.

Component

Details

Operating system

Gaia OS

Service

DynamicCLI daemon

Process privilege

Root

IPC mechanism

Local Unix socket

Socket path

/tmp/dynamic_socket

Permission issue

World-writable socket

Authentication issue

Connecting processes are not properly authenticated

Exploitation result

Command execution as root

The daemon's elevated privileges make the socket an attractive target.

An attacker does not need to exploit a separate kernel vulnerability to obtain elevated privileges through this path. The issue lies in the way a privileged service exposes and processes local IPC requests.

Forensic significance

Incident responders should investigate:

  • Unexpected commands executed through DynamicCLI.

  • Suspicious process execution associated with dynamic_server.

  • Unusual use of the system() command-execution sink.

  • Commands that bypass the normal Clish wrapper.

  • Unexpected webshells and backdoor files.

  • TSGA and TSProxy-related artifacts.

  • Suspicious activity involving VMware ESXi infrastructure.

Check Point's reported incident activity also demonstrates why privilege-escalation vulnerabilities in security appliances deserve particular attention: compromise of a gateway or management appliance can create opportunities for further access into the protected environment.

UnrealIRCd: Labeled-Response Module Vulnerability

The UnrealIRCd vulnerability involves improper handling of a protocol message tag.

The affected component is the labeled-response module, which processes IRCv3 label tags and includes them in responses.

The failure to escape the label correctly creates an opportunity for malicious content to be interpreted as additional server-to-server commands.

Component

Details

Application

UnrealIRCd

Affected module

labeled-response

Protocol feature

IRCv3 message tags

Vulnerable field

label

Weakness

Improper delimiter neutralization

Network requirement

Two or more linked IRC servers

Exploitation result

Commands execute with server authority

Forensic significance

IRC administrators should review:

  • Unexpected privileged IRC commands.

  • Unusual KILL and GLINE operations.

  • Suspicious server-to-server messages.

  • Unexpected changes to IRC network configuration.

  • Unusual user activity immediately preceding administrative commands.

  • Whether every linked server has received the security fix.

The UnrealIRCd maintainers also provided a method to verify whether the labeled-response hot-patch is installed.

Indicators of Compromise (IOCs)

The CISA alert does not publish a dedicated list of malicious IP addresses, domains or file hashes for these two vulnerabilities.

The following are technical investigation indicators derived from the vulnerability descriptions and vendor guidance, not confirmed network IOCs.

Check Point

Indicator

Description

/tmp/dynamic_socket

DynamicCLI IPC socket associated with the vulnerability

dynamic_server

Root-level Gaia OS DynamicCLI daemon

Unexpected root commands

Commands executed through the vulnerable IPC interface

Suspicious system() usage

Potential indication of command execution requiring investigation

TSGA

Backdoor referenced in related Check Point exploitation reporting

TSProxy

Malware referenced in related Check Point exploitation reporting

Unexpected webshells

Reported post-exploitation artifact

UnrealIRCd

Indicator

Description

labeled-response

Affected UnrealIRCd module

Malformed IRCv3 label tags

Potential exploitation input

Unexpected KILL commands

Possible unauthorized IRC operator activity

Unexpected GLINE commands

Possible unauthorized network administration

Suspicious server-to-server commands

Potential command-injection activity

IOC handling note: These indicators are intended to support investigation. Their presence alone does not prove exploitation. The CISA alert does not provide confirmed malicious hashes, attacker IP addresses or command payload samples.

Potential Impact

1. Root-Level System Compromise

CVE-2026-123039 can allow an authenticated local attacker to execute commands with root privileges on affected Check Point appliances.

2. Security Appliance Compromise

A compromised Check Point gateway or management server may provide an attacker with a foothold in critical network infrastructure.

3. Persistence and Lateral Movement

Related Check Point incident reporting describes webshells, backdoor deployment and movement toward VMware ESXi infrastructure.

4. IRC Network Administration Abuse

CVE-2026-114713 can allow ordinary users to execute commands with server authority on vulnerable linked IRC networks.

5. Service Disruption

Unauthorized administrative commands may affect users, disrupt network operations or interfere with IRC services.

6. Destructive Activity

The Check Point exploitation reporting describes a subsequent wiper deployment against VMware ESXi infrastructure. This demonstrates the potential consequences of a broader compromise, although it should not be assumed that every exploitation of CVE-2026-123039 results in a wiper attack.

Recommendations

Immediate Actions

For Check Point administrators

  1. Identify all deployed Check Point products included in the vendor's affected-product guidance.

  2. Review the applicable software versions and Jumbo Hotfix levels.

  3. Apply the emergency hotfixes and mitigations specified in Check Point Security Advisory SK1000362.

  4. Investigate suspicious DynamicCLI activity and unexpected root-level commands.

  5. Review authentication records, process execution and system logs for signs of unauthorized access.

  6. Examine appliances for webshells, TSGA, TSProxy and other suspicious persistence artifacts.

  7. Investigate connected infrastructure, including VMware ESXi systems, if evidence indicates post-exploitation movement.

For UnrealIRCd administrators

  1. Upgrade vulnerable installations to UnrealIRCd 6.2.7.1 or apply the official hot-patch.

  2. Verify that the labeled-response module is patched on every server in the network.

  3. Review IRC logs for unexpected privileged commands and suspicious server-to-server activity.

  4. Investigate unauthorized KILL and GLINE operations.

  5. Avoid continuing to operate vulnerable end-of-life UnrealIRCd versions.

Preventive Actions

  1. Reduce local privilege exposure: Restrict access to privileged IPC interfaces and ensure that socket permissions and authentication controls are correctly configured.

  2. Apply least privilege: Limit local access to security appliances and administrative interfaces.

  3. Monitor security appliances: Centralize logs from gateways and management servers and alert on unusual privileged command execution.

  4. Harden IRC infrastructure: Keep UnrealIRCd updated and review linked-server configurations.

  5. Verify patches: Confirm the installed hotfix or patched module rather than relying solely on update deployment records.

  6. Perform forensic triage: Treat potentially compromised systems as security incidents, even if a patch has already been installed.

  7. Maintain network segmentation: Restrict access between security appliances, management infrastructure and critical systems.

CISA KEV Catalog and Compliance Requirements

Both vulnerabilities were added to the KEV Catalog on November 13, 2026.

Field

CVE-2026-123039

CVE-2026-114713

Vendor

Check Point

UnrealIRCd

Vulnerability

Incorrect permission assignment

Improper neutralization of delimiters

Affected component

Gaia OS DynamicCLI

Labeled-response module

Exploitation impact

Local privilege escalation to root

Server-authority command injection

KEV addition date

November 13, 2026

November 13, 2026

Federal remediation deadline

November 16, 2026

November 16, 2026

Known ransomware campaign use

Unknown

Unknown

Forensic triage required

Yes

Yes

Binding Operational Directive 26-04

CISA referenced Binding Operational Directive (BOD) 26-04, which establishes vulnerability-management requirements for Federal Civilian Executive Branch (FCEB) agencies.

The directive reinforces the importance of prioritizing KEV-listed vulnerabilities, particularly vulnerabilities on publicly exposed assets that can grant total control after exploitation.

It also establishes expectations for checking whether threat actors compromised systems before patches were applied.

Although the directive applies to FCEB agencies, CISA encourages all organizations to use the KEV Catalog as an input to risk-based vulnerability management.

Important: The November 16 deadline is the federal remediation deadline. It should not be interpreted as a universal legal deadline for every private organization.

Conclusion

CISA's November 13, 2026 KEV update highlights two vulnerabilities that affect different but important areas of infrastructure security.

CVE-2026-123039 exposes Check Point appliances to local privilege escalation through an improperly protected DynamicCLI socket. CVE-2026-114713 allows command injection in UnrealIRCd networks through improper handling of IRCv3 label tags.

Both vulnerabilities have evidence of active exploitation and require urgent attention from administrators operating affected products.

The Check Point incident reporting demonstrates how compromise of a security appliance can lead to persistence, further network access and destructive activity. Meanwhile, the UnrealIRCd vulnerability illustrates how improper protocol-message handling can undermine the administrative security of linked server networks.

Organizations should prioritize vendor fixes, validate patch deployment, monitor suspicious activity and perform forensic triage to determine whether exploitation occurred before remediation.

Patching closes the vulnerability. Forensic triage helps establish whether the attacker already used it.

MITRE ATT&CK Mapping

T1068 — Exploitation for Privilege EscalationT1059 — Command and Scripting InterpreterT1046 — Network Service Discovery