Remote Desktop Client CVE-2026-42992

Exploitation requires the victim to initiate or accept an RDP connection from the vulnerable Remote Desktop Client to an attacker-controlled RDP server (UI:R), giving the malicious server the opportunity to send the crafted PDU that triggers the heap overflow; no credentials on the client side are needed (PR:N). Exploitation is limited by AC:H, meaning the attacker must satisfy an additional condition beyond simply serving a malformed packet - likely winning a race or successfully grooming the client heap - and by the requirement that the user be willing to connect outbound to an untrusted RDP endpoint, which is uncommon in hardened enterprise environments that restrict outbound 3389/443-RDP traffic. The flaw cannot be triggered server-to-client without that outbound connection, so listening RDP servers (the usual RDP attack surface) are not the vulnerable component here.

The CVSS 7.5 vector (AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H) indicates a network-reachable, unauthenticated bug with high impact but elevated attack complexity and required user interaction - meaning a victim must initiate or accept an RDP session to a malicious server, and the attacker must satisfy some non-trivial condition (likely a heap-grooming or race window) to achieve reliable exploitation. There is no public exploit identified at time of analysis, the CVE is not on CISA KEV, and no EPSS score was supplied in the input, so quantitative exploitation probability cannot be cited. Real-world risk is highest for environments where users routinely connect outbound to third-party RDP gateways, jump hosts, or hosted desktops; in tightly-scoped internal-only RDP deployments the realistic exposure is significantly lower than the raw CVSS suggests.

An attacker stands up a malicious RDP server and lures a target - via phishing with a crafted .rdp file, a typo-squatted hosted-desktop link, or a compromised RD Gateway - into initiating a connection. During session negotiation the server returns a malformed PDU that overflows a heap buffer in the client parser, leading to code execution in the user's context on the victim workstation. No public exploit is identified at time of analysis, but the AC:H rating implies a working exploit would require heap shaping rather than being impossible.