Cybersecurity researchers have revealed that threat actors swiftly exploited a critical remote code execution vulnerability in Cisco Expressway gateways to successfully breach a Russian organization. The attack materialized almost immediately after proof-of-concept (PoC) exploit code was published on public platforms, highlighting the persistent danger of rapid weaponization that modern security teams face within hours of vulnerability disclosures.
The security flaw, formally tracked as CVE-2026-76466 and carrying a maximum severity CVSS score of 9.8, affects both virtual and appliance-based deployments of Cisco Expressway Core and Expressway Edge. However, the vulnerability’s impact is specifically scoped to environments where these devices are deployed in clusters. Expressway gateways serve as critical enterprise infrastructure, allowing remote employees, trusted partners, and third-party users to securely connect to an organization’s essential collaboration services, including voice calls, video conferences, and instant messaging solutions.
Cisco originally disclosed the vulnerability on October 21, explaining that the issue resides within the database service feature of the platform and stems from improper handling of persistent cookies. According to the technical advisory provided by the networking giant, an unauthenticated remote attacker with network access to a vulnerable device can leverage these persistent cookies to reach the database service. Once inside, the adversary can issue specific remote procedure call (RPC) commands, effectively granting them the capability to execute arbitrary commands with full root privileges on the underlying operating system.
Despite the critical nature of the flaw, Cisco’s official security advisory has remained unchanged since its initial publication and still notes that the company is unaware of any public announcements or active malicious exploitation of the vulnerability. Nevertheless, the cybersecurity landscape shifted dramatically when a functional PoC exploit script was posted publicly on GitHub under the identifier associated with the CVE. According to findings published by cybersecurity firm F6, the real-world attack they subsequently investigated took place in close temporal proximity to the release of this public exploit code.
In the incident analyzed by F6, the attackers initiated their campaign by targeting two internet-exposed Cisco Expressway X15.5.3 devices. After successfully compromising these initial edge gateways, the threat actors pivoted internally, moving from one of the breached perimeter nodes to two additional cluster nodes operating securely within the targeted organization’s internal network architecture.
Having secured high-level root privileges across the compromised gateway infrastructure, the attackers immediately began preparing a suite of tools designed to facilitate lateral movement across the broader corporate network. F6 reported that the recovered toolset included custom Python scripts engineered to autonomously scan the internal network, locate vulnerable Server Message Block (SMB) resources, interact directly with SQL Server and Windows Remote Management (WinRM), and search internal GitLab code repositories. The primary objective of these reconnaissance and traversal efforts appeared to be the unauthorized exfiltration of credentials and other sensitive organizational data.
A particularly notable discovery during F6’s forensic investigation was that the attackers actively modified and optimized their malicious scripts on the fly while the operation was actively underway. The security firm noted that the underlying source code contained distinct characteristic comments and explicit descriptions of bug fixes that strongly indicated the use of an artificial intelligence assistant by the threat actors during the attack lifecycle.
F6 has chosen not to publicly name the victim organization targeted in the incident, nor has the firm attributed the attack to any specific, previously known advanced persistent threat group or threat actor collective.
To help security teams and enterprise defenders identify potential exploitation attempts within their own environments, F6 pointed administrators toward the Erlang system logs generated on Cisco Expressway devices. Furthermore, the security company has published specific indicators of compromise, including malicious IP addresses and suspicious file paths associated with the observed attack, as part of a detailed Russian-language press release.
Because Cisco has clarified that there are no immediate software workarounds available to mitigate the risk aside from upgrading, organizations operating vulnerable deployments are under immense pressure. Customers currently running Expressway X14 or earlier legacy releases must upgrade directly to a fixed software release, specifically Expressway X15.5.20, which fully resolves the underlying security defect. For organizations that are unable to apply the necessary software patches immediately, F6 has recommended a temporary defensive measure: blocking all network connections to TCP ports 4371 through 4372 at the network firewall level, with the sole exception of trusted traffic originating from the IP addresses of verified cluster nodes.
SecurityWeek has reached out to Cisco for formal comment regarding F6’s detailed findings concerning the active exploitation of CVE-2026-76466 in the wild and will provide updates should the networking vendor issue a response.

