Mastodon Mastodon Mastodon Mastodon

Critical UnrealIRCd vulnerability allows commands to be executed with server privileges — exploitation already underway

Photo of author

CyberSecureFox Editorial Team

Published:

On November 13, 2026, CISA added the CVE-2026-114713 vulnerability to its Known Exploited Vulnerabilities (KEV) catalog following reports of active exploitation. The vulnerability affects UnrealIRCd — a popular open‑source IRC server — and allows a regular, non‑privileged user to inject commands that are executed with server or IRC operator privileges. The CVSS 3.1 score is 8.6 (High). Versions from 6.0.0 through 6.2.7 inclusive are affected; a fix is available in version 6.2.7.1 and as a hot patch. U.S. federal agencies have been given a remediation deadline of November 16, 2026.

Technical essence of the vulnerability

The issue belongs to the class CWE-140 — Improper Neutralization of Delimiters. The labeled-response module implements an IRCv3 mechanism in which the server echoes back to the client the value of the label tag in response messages to correlate requests and responses. As stated in the NVD description, the module did not escape delimiter characters in the tag value before writing it into the outgoing message. This allows an attacker to inject arbitrary inter‑server protocol commands, which are then executed with server privileges.

According to a technical analysis, the actions available to an attacker include such privileged commands as KILL (forcibly disconnecting a user) and GLINE (global ban), which effectively gives an ordinary user the capabilities of a network administrator.

Exploitation conditions: not all configurations are equally vulnerable

An important detail to consider when assessing risk: full privilege escalation to the server level requires two conditions to be met simultaneously — the labeled-response module must be loaded, and the network must contain at least two linked UnrealIRCd servers. On a standalone instance with a single server, this particular privilege escalation vector does not work, although the protocol handling is still incorrect and needs to be fixed. Thus, large IRC networks with multiple linked servers are at the highest risk.

Active exploitation and timeline

According to a statement by project maintainer Bram Matthys (Syzop) published on the UnrealIRCd forum, attempts to abuse the issue began the day after the fix was released on November 7, 2026. Matthys described the vulnerability as “the most critical issue” found in UnrealIRCd over the past 15 years. At the same time, the project labels the vulnerability as critical, while the numerical CVSS score is 8.6 (High) — this discrepancy should be taken into account when prioritizing.

The attackers’ specific objectives, scale of exploitation, victims, and links to any known groups are currently unknown. Available information is limited to the vendor’s statements and the status in the CISA KEV catalog; no independent telemetry or victim reports have been found in open sources.

Affected versions

  • UnrealIRCd 6.0.0 — 6.2.7 — vulnerable. Version 6.2.7 is not fixed: you must update to 6.2.7.1 or apply the hot patch.
  • UnrealIRCd 5.x — also affected, but this branch reached End of Life on July 1, 2023. No fixed version has been released for it. The vendor recommends disabling the labeled-response module and migrating to a supported version.

Remediation recommendations

According to the vendor’s guidance, the following remediation options are available:

  1. Hot patch (no restart) — bundle-20261107, applicable to versions 6.0.0–6.2.7. Does not require a server restart, but according to the vendor is not available for Windows installations.
  2. Full upgrade to version 6.2.7.1 — with a server restart.
  3. Temporary SPAMFILTER workaround — protects only against the critical vulnerability and only if all linked servers in the network are running version 6.0.4 or newer. This workaround does not address the other eight security issues fixed in release 6.2.7.1.

Patch status verification

The verification nuance deserves special attention: remote version checks via the server banner are insufficient to confirm that the vulnerability has been fixed. A server with the hot patch applied will still display the version UnrealIRCd-6.2.7 in the banner, making it impossible to distinguish a fixed system from a vulnerable one remotely. Reliable verification requires using the local command ./unrealircd module-status or the IRC MODULE command with operator privileges:

  • labeled-response 5.0-hot-patch-bundle-20261107 — the hot patch is applied, the server is protected.
  • Module version 6.2.7.1 — the full fixed version is installed.
  • Value 5.0 without a patch suffix — the hot patch is not applied, the server is vulnerable.

This detail is especially important for teams auditing infrastructure: banner scanning will produce false negatives on hot‑patched servers and cannot be used as a basis for concluding that the system is secure.

Impact assessment

Although IRC has long given way to more modern messaging platforms, UnrealIRCd remains one of the most widespread IRC servers and is still used by a number of communities and organizations. The ability to execute commands with server privileges means an attacker can disconnect users, impose global bans, and potentially disrupt the operation of the entire IRC network — all without any special privileges, acting as a regular connected user.

UnrealIRCd administrators should immediately update to version 6.2.7.1 or apply the bundle-20261107 hot patch, and then verify the state of the labeled-response module via module-status. Operators of networks running UnrealIRCd 5 must disable the labeled-response module and begin migrating to a supported branch. The SPAMFILTER workaround is only a temporary measure and is not a substitute for a full upgrade.


CyberSecureFox Editorial Team

The CyberSecureFox Editorial Team covers cybersecurity news, vulnerabilities, malware campaigns, ransomware activity, AI security, cloud security, and vendor security advisories. Articles are prepared using official advisories, CVE/NVD data, CISA alerts, vendor publications, and public research reports. Content is reviewed before publication and updated when new information becomes available.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.