The vulnerability is a command injection flaw present in the HyperV machine backend of Podman. In versions 4.8.0 through 5.8.1, the VM image path is concatenated directly into a double-quoted PowerShell command without sanitization. An attacker who can influence the VM image path—for example by providing a specially crafted machine name or specifying an image directory—can inject a $() subexpression. PowerShell evaluates that subexpression before executing the outer command, which allows arbitrary PowerShell code to run with the same privileges as the Podman process. On Windows this means potential SYSTEM-level execution.

The affected product is Podman, the container management tool provided by the containers:podman project. Versions 4.8.0 through 5.8.1 are vulnerable; all other releases are unaffected. The issue is exclusive to the HyperV backend and therefore only affects Windows installations. The vulnerability has been fixed in Podman 5.8.2.

The CVSS score of 4 indicates a low severity, yet the ability to execute code at the Podman process level turns this into a remote code execution risk. An EPSS score of <1% (approximately 0.00028) shows a low but non‑zero likelihood of exploitation. The vulnerability has not been listed in CISA's KEV catalog. Attackers need to control the VM image path, which can be done if they can supply a machine name or image directory during machine creation, typically via local or remote access to the Podman service. Because the flaw uses PowerShell injection, the attack vector is Windows‑only, and the impact is equivalent to running arbitrary commands with SYSTEM privileges if Podman is elevated.