Palo Alto Issues Highest Urgency Rating for Unauthenticated PAN-OS Firewall Flaw

Palo Alto Networks has disclosed a set of buffer overflow vulnerabilities in the User-ID Terminal Server Agent component of PAN-OS, assigning the flaw its own highest possible urgency rating. Tracked as CVE-2026-0288, the bug carries a CVSS-B score of 9.2 and can be triggered by an unauthenticated attacker simply by reaching the Terminal Server Agent's IP address and port over the network.

The root cause is a CWE-787 out-of-bounds write in how the Terminal Server Agent processes incoming network traffic. An attacker with network access to that agent can send specially crafted packets that corrupt memory, with the potential to trigger a denial of service or, in the worst case, achieve arbitrary code execution. No authentication and no user interaction are required, which is precisely the combination that turns a serious bug into an urgent one.

Not every deployment is exposed equally. Palo Alto Networks confirmed that Panorama is unaffected, and the flaw only applies to devices with at least one Terminal Server Agent entry configured under Device, User Identification, Terminal Server Agents. Severity also shifts sharply with exposure. Devices with Terminal Server Agent access reachable from the internet or untrusted networks sit at the full 9.2 rating, while agents restricted to trusted internal segments drop to 7.7. Prisma Access customers carry comparatively lower risk since exploitation there requires an authenticated user and external Terminal Server Agent access is already restricted by default.

Affected versions and fixes

The vulnerability spans the PAN-OS 10.2, 11.1, 11.2 and 12.1 branches, along with Prisma Access 10.2 and 11.2, where it is rated medium severity. Cloud NGFW on AWS is unaffected, and Cloud NGFW on Azure customers should check directly with Palo Alto Networks regarding their specific exposure. Fixed builds are available across every affected branch, and the vendor is urging organisations to upgrade rather than wait, given the network-based attack vector and low exploitation complexity. Palo Alto Networks has not reported evidence of active exploitation at the time of disclosure, but a vendor's own highest urgency label has repeatedly preceded confirmed in-the-wild exploitation within days rather than weeks in comparable cases this year, a pattern worth weighing more heavily than the absence of a mandated compliance deadline.

Why GCC enterprises should treat this as immediate

Palo Alto firewalls are among the most widely deployed next generation firewall platforms across GCC government, financial services and critical infrastructure environments, which makes this advisory relevant well beyond the vendor's own customer base considerations. Any organisation running an exposed User-ID deployment should not wait for a compliance deadline to act, since the absence of a mandated remediation window does not reflect the absence of risk, only the absence of a regulator with jurisdiction to impose one. The same logic applies broadly across vendor security disclosures, a point MCW has raised before in the context of source code level compromises at security vendors themselves, where the absence of confirmed customer impact at disclosure time did not mean the risk to downstream enterprises was low.

As an interim mitigation ahead of patching, Palo Alto Networks recommends restricting Terminal Server Agent connectivity to trusted internal IP addresses only, in line with its published best practice deployment guidance. This step alone meaningfully reduces the attack surface even before an upgrade is scheduled. Security teams should also treat this disclosure as a prompt to audit every Terminal Server Agent entry currently configured across their firewall estate, since the flaw only manifests where that specific integration is present rather than across all PAN-OS deployments broadly.

The vulnerability was responsibly disclosed by external researcher Liang Zhu, continuing a pattern this year of independent researchers identifying high severity flaws in widely deployed network security infrastructure well ahead of any confirmed exploitation.