JetBrains TeamCity EUVD-2026-33381

Exploitation requires the TeamCity instance to use a Perforce (Helix Core) version-control connection, and the attacker must be an authenticated user with sufficient privileges (PR:L) to create or modify that Perforce connection/VCS-root configuration - the malicious input is delivered through the Perforce connection settings themselves. Limiting factors: the attack is not possible against instances that do not use Perforce, and it cannot be performed by fully unauthenticated outsiders because it depends on access to configuration that TeamCity normally restricts to project administrators; no user interaction from a victim is needed once the attacker can edit the settings.

The signals are mixed and should be weighed carefully rather than taken from the 8.8 headline alone. The CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) shows network reachability, low complexity, no user interaction, and full confidentiality/integrity/availability impact, but critically requires privileges (PR:L) - the attacker must already be an authenticated TeamCity user able to view or edit Perforce VCS root/connection settings, which is a meaningful gate in most deployments. This is corroborated by SSVC, which rates Exploitation as none, Automatable as no, and Technical Impact as total: high consequence if triggered, but not a spray-and-pray target. EPSS is 0.00% and the CVE is not in CISA KEV, so there is no evidence of active or imminent mass exploitation. Net assessment: a genuine high-impact RCE that matters most where untrusted or lower-trust users hold project-configuration rights, but not an internet-wide unauthenticated emergency. Note one limitation to verify: the description does not spell out the exact privilege level needed to modify Perforce settings, so the PR:L assumption should be confirmed against your TeamCity role model.

An attacker who holds a TeamCity account with rights to edit a project's Perforce VCS connection injects crafted argument delimiters into the p4 connection parameters; when TeamCity next invokes the Perforce client, the smuggled arguments execute attacker-controlled code on the build server. Given AC:L and no user interaction required, the action succeeds reliably once the malicious settings are saved, though no public proof-of-concept has been identified.