Proxmox Mail Gateway CVE-2026-94588

Exploitation requires the attacker to reach the package changelog retrieval functionality of Proxmox pmg-api with a valid authenticated (per CVSS PR:H, administrative) session, OR to trick an already-authenticated PMG administrator into triggering the request CSRF-style from their browser. The specific vulnerable feature is the changelog-fetch path that passes user-supplied input into the underlying apt-get command without sanitization. Limiting factors: the target must be a Proxmox Mail Gateway deployment (niche), an admin must be authenticated, and for the CSRF variant the admin must interact with attacker-controlled content - consistent with the high attack complexity (AC:H).

This is a low-to-moderate real-world risk despite the network attack vector. The CVSS 3.1 vector (AV:N/AC:H/PR:H/UI:N/S:C/C:L/I:L/A:N) yields a modest 4.4 score, driven down by high attack complexity (AC:H), the requirement for high privileges (PR:H), and only Low confidentiality/integrity impact with no availability impact. No EPSS score, no CISA KEV listing, and no public POC are present in the provided data, so exploitation status is: no public exploit identified at time of analysis. The description states authentication is required but exploitation can occur CSRF-style, which conflicts with the vendor's UI:N - a genuine CSRF attack presupposes victim user interaction (a logged-in admin's browser being tricked), so UI:N should likely be UI:R; verify with vendor. The single reference is a Proxmox forum advisory with no enumerated affected versions, no patch version, and no independent corroboration, so the affected-version scope is unconfirmed. Overall this reads as a real-but-low-priority hardening issue in a niche management API rather than a widely-weaponizable flaw.

An attacker crafts a malicious web page or link that, when visited by an authenticated Proxmox Mail Gateway administrator, silently issues a package-changelog request to pmg-api containing attacker-controlled apt-get arguments (CSRF-style). The injected arguments are passed unsanitized to apt-get running at elevated privilege, altering its behavior. No public exploit code is identified at time of analysis, and the high attack complexity plus need for an active admin session limit reliability.