Multicluster Engine EUVD-2026-92524

Exploitation requires an authenticated user who holds both cluster-creation and secret-creation permissions in a HyperShift/Multicluster Engine deployment; the attacker must supply a user-provided kubeconfig secret containing an unauthorized executable credential plugin (exec-based auth), which the HyperShift operator copies unvalidated into the privileged control plane namespace. Code execution only fires when a downstream controller actually consumes that kubeconfig. Limiting factors: no anonymous access (valid account with the specific RBAC grants is mandatory, per PR:L), and the environment must be running HyperShift hosted control planes - this is not exploitable against a product without that operator deployed.

CVSS 3.1 scores this 8.8 (High) with vector AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, indicating network-reachable, low-complexity exploitation requiring only low privileges and no user interaction, with full confidentiality/integrity/availability impact. The key gating signal is PR:L: this is NOT an anonymous remote RCE - the description confirms the attacker must already be an authenticated user holding cluster-creation and secret-creation RBAC permissions, which meaningfully narrows the realistic attacker population to tenants/operators already inside the management environment. No EPSS score, no CISA KEV listing, and no public POC were provided in the input, so there is no evidence of active exploitation or weaponization; the severity rests on the CVSS impact metrics rather than observed in-the-wild activity. Notably, this is a privilege-escalation/tenant-escape pattern (lower-privileged user achieving code execution in the privileged control plane namespace) rather than a mass-exploitable internet-facing bug - that argues for CVSS scope change (S:C), yet Red Hat rated S:U, a point worth verifying with the vendor. Overall this is a genuine high-priority issue for organizations running HyperShift hosted control planes due to the control-plane RCE outcome, but the authentication prerequisite makes it an insider/multi-tenant-abuse risk rather than a drive-by internet threat.

An authenticated tenant with permission to create a hosted cluster and its secrets submits a kubeconfig secret containing a malicious exec credential plugin (an arbitrary executable/command). The HyperShift operator copies this secret unvalidated into the privileged control plane namespace, and when a downstream controller consumes the kubeconfig it invokes the attacker's plugin, yielding arbitrary code execution inside the control plane. No public exploit identified at time of analysis, but the technique is straightforward given the required permissions.