CVE-2026-129120 in C-Kermit
Summary
by MITRE • 11/13/2026
C-Kermit 11.0.506 before 11.1.515 contains a RECEIVE CONFIRM bypass vulnerability that allows malicious Kermit servers to write unrequested files without prompting by abusing recursive GET root computation in ckcfns.c. Attackers can send files such as .ssh/authorized_keys or shell startup files under the current directory, which rq_under_root() accepts silently, enabling persistent access or code execution.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 11/13/2026
The vulnerability identified in C-Kermit versions prior to 11.1.515 represents a critical security flaw within the file transfer protocol implementation, specifically targeting the RECEIVE CONFIRM mechanism and root directory validation logic. C-Kermit is a widely used program for transferring files between different computer systems using various protocols, including Kermit, XMODEM, ZMODEM, and others. The core of this vulnerability lies in the handling of recursive GET operations within the source file ckcfns.c. When a client initiates a transfer from a server that has been compromised or controlled by an attacker, the application fails to properly validate whether the requested files are located outside the designated root directory or if they bypass intended access controls. This flaw allows malicious servers to trick the C-Kermit client into downloading and writing unrequested files directly to sensitive locations on the victim's system without triggering standard confirmation prompts that would normally alert the user to unexpected file operations.
From a technical perspective, the exploit relies on abusing the recursive GET root computation logic. The function rq_under_root() is designed to ensure that transferred files remain within an acceptable directory structure, preventing path traversal attacks where attackers might attempt to write files outside of allowed directories using sequences like ../.. However, due to insufficient validation in versions before 11.1.515, this check can be bypassed. An attacker controlling a Kermit server can craft specific file paths or transfer requests that exploit the recursive nature of the GET command. By doing so, they can force the client to write files such as .ssh/authorized_keys for SSH key injection or shell startup scripts like .bashrc or .profile in the current working directory. These locations are particularly dangerous because modifications here often lead directly to persistent access or arbitrary code execution upon subsequent user interactions or system events.
The operational impact of this vulnerability is severe, primarily facilitating unauthorized persistence and remote code execution on targeted systems. If an attacker successfully writes a public key to .ssh/authorized_keys, they gain immediate SSH access without needing the victim's password, effectively establishing a backdoor that persists across reboots and user logouts. Similarly, injecting malicious commands into shell startup files ensures that any time a new terminal session is opened or a script runs in that directory, the attacker’s code executes with the privileges of the current user. This capability transforms what might otherwise be a simple file transfer utility into a potent vector for initial access and lateral movement within an organization's network infrastructure. The lack of prompting means victims are unlikely to notice these unauthorized writes until significant damage has been done or persistent access is established, complicating incident response efforts.
This vulnerability aligns with several well-known industry standards regarding software weaknesses and attack patterns. It corresponds closely to CWE-20 Improper Input Validation, as the application fails to adequately sanitize or verify input related to file paths during transfer operations. Additionally, it relates to CWE-73 External Control of File Name or Path, where an attacker influences the location where files are written on a system. In terms of offensive security frameworks, this behavior mirrors techniques found in MITRE ATT&CK under T1505.002 Server Software Component: Web Shell, although adapted for file transfer protocols rather than web servers specifically, and potentially T1078 Valid Accounts if used to establish unauthorized access credentials. The abuse of legitimate administrative tools like C-Kermit also reflects the broader category of Living off the Land binaries (LOLBins), where attackers leverage trusted software to evade detection by security monitoring solutions that typically flag malicious executables but not benign utilities performing suspicious file operations.
Mitigation strategies for this vulnerability are straightforward and primarily involve updating the affected software. Organizations relying on C-Kermit should immediately upgrade to version 11.1.515 or later, where the developers have addressed the flaws in rq_under_root() and reinforced the RECEIVE CONFIRM bypass protections. For environments where upgrading is not immediately feasible, strict network segmentation can help mitigate risk by restricting which systems are allowed to initiate Kermit transfers with internal clients. Furthermore, implementing file integrity monitoring on critical directories such as .ssh/ or home directory root locations can provide early detection of unauthorized changes. Security awareness training should also be emphasized, reminding users that even trusted utilities like C-Kermit must only connect to verified and secure servers, as the trust model assumes the remote party is benign until proven otherwise. Regular audits of installed software versions and prompt patching remain the most effective defense against such protocol-level exploitation attempts.