A logic flaw in the F Prime FileUplink component causes an unsigned integer overflow when calculating the boundary byteOffset + dataSize against the fileSize. An attacker can send a crafted packet with a large byteOffset that wraps to zero, bypassing the bounds check and writing data starting at an offset close to the full 4GB range. In addition, the code that writes to the destination file performs no validation of the file path, enabling a write to any arbitrary file. The combination of these defects permits an attacker to inject arbitrary data into any file on the target system, which can then be used to achieve remote code execution on the embedded device. Affected systems are deployments of NASA’s F Prime framework older than version 4.2.0. The vulnerability is present in the natural file uplink used by many spaceflight and other embedded software applications that rely on F Prime for data transfer. Risk and exploitability: The flaw has no public exploit probability score available and is not listed in the CISA KEV catalog, but its potential impact is severe. The vulnerability requires the ability to send a specially crafted data packet and access to the FileUplink service, which may be exposed locally or remotely depending on the deployment. Because the corruption occurs at the file I/O layer, typical memory corruption detection tools do not catch it. Attackers who succeed can overwrite critical binaries or configuration files, resulting in complete compromise of an embedded target.

Affected systems are deployments of NASA’s F Prime framework older than version 4.2.0. The vulnerability is present in the natural file uplink used by many spaceflight and other embedded software applications that rely on F Prime for data transfer.

Risk and exploitability: The flaw has no public exploit probability score available and is not listed in the CISA KEV catalog, but its potential impact is severe. The vulnerability requires the ability to send a specially crafted data packet and access to the FileUplink service, which may be exposed locally or remotely depending on the deployment. Because the corruption occurs at the file I/O layer, typical memory corruption detection tools do not catch it. Attackers who succeed can overwrite critical binaries or configuration files, resulting in complete compromise of an embedded target.