Forcepoint VPN Client CVE-2025-12694

Attacker must already have local interactive or remote-shell access as an authenticated low-privileged Windows user (PR:L) on a host where Forcepoint VPN Client for Windows 6.11.3 or earlier is installed and its privileged service is running; no user interaction by another party is required (UI:N) and attack complexity is low (AC:L). Exploitation is limited by the local-only attack vector (AV:L) - it cannot be triggered remotely over the network - and by the requirement that the vulnerable client be present, so endpoints without Forcepoint VPN Client deployed are not exposed regardless of network position.

Risk is moderate-to-high for endpoints where the Forcepoint VPN Client is deployed: CVSS 4.0 base of 8.5 reflects local attack vector (AV:L), low complexity (AC:L), low privileges required (PR:L), no user interaction (UI:N), and full host impact, which is a realistic LPE profile for a privileged endpoint agent. No EPSS score, KEV listing, or public POC was supplied in the intelligence, so this is not a mass-exploitation candidate today - it is more relevant to insider-threat, malware post-compromise, and red-team scenarios where SYSTEM is the goal after an initial foothold. The advisory originated from Forcepoint PSIRT, suggesting coordinated disclosure rather than in-the-wild discovery; however, absence of public exploit code does not preclude private weaponization given how attractive an LPE in a security-product agent is for adversaries chaining out of low-privilege footholds.

An attacker who has already obtained a standard Windows user session - through phishing, a stolen password, or a separate browser exploit - invokes a crafted interaction with the privileged Forcepoint VPN Client component (e.g., a service, scheduled task, or IPC endpoint) to coerce it into running attacker-controlled code as SYSTEM. The result is full host takeover, allowing credential theft via LSASS, persistence installation, and lateral movement from a previously unprivileged foothold. No public exploit identified at time of analysis, but CWE-250 LPEs in endpoint agents are routinely reverse-engineered from patch diffs.