CVE-2026-110930
Description
A flaw was found in Codex Desktop. Improper validation of quotation marks in project paths allows an attacker to inject application startup arguments when a user opens an attacker-controlled project through Windows Explorer. This issue can lead to arbitrary code execution with the privileges of the logged-in user before user sign-in or project trust verification occurs.
Statement
Red Hat Product Security rates this vulnerability as having an Important impact because successful exploitation enables local code execution under the active user account without requiring pre-authentication. Red Hat products are not affected by this flaw, as Codex Desktop is a Windows-specific application that is not packaged, supported, or distributed in any Red Hat offering.
Mitigation
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
--- CME Mitigation Guidance (2026-11-06) ---
Disclaimer: The following mitigation solutions are provided as a courtesy, pending the application of an official patch. This content was generated utilizing AI tools in conjunction with data from https://cmetaxonomy.org. Customers are advised to independently evaluate and test these temporary mitigations within their own infrastructure.
The highest-value action is to sanitize unescaped quotation marks and delimiter characters in project paths passed to the application command line to prevent startup argument injection: upgrade to Codex Desktop 26.1007.21434, which validates and sanitizes quotation marks and delimiter characters in project paths received from Windows File Explorer before passing them to the application command line, closing all 2 injection vectors (crafted project paths with quotation mark injection, malicious .library-ms context menu invocation). Until patched, this is only a concern on Windows systems running Codex Desktop — Linux, macOS, and Red Hat platforms are unaffected — so Avoid opening untrusted .library-ms files or invoking 'Open project in ChatGPT' from the Windows File Explorer context menu while Codex Desktop is closed. As a compensating measure, Deploy Windows Defender Application Control in enforced mode to block unauthorized binaries or unsigned scripts triggered as downstream payloads and Configure an EDR agent to detect anomalous process execution, unusual command-line arguments, and suspicious child processes. These controls limit blast radius and contain exploitation, but they do not fix the underlying argument injection defect — apply the upstream patch to fully remediate.
Highest-impact CME: CME-1312 Argument Injection Prevention (Downstream Input Delimiter Sanitization) (high)
Understanding the Weakness (CWE)
Confidentiality,Integrity,Availability,Other
Technical Impact: Execute Unauthorized Code or Commands; Alter Execution Logic; Read Application Data; Modify Application Data
If an attacker can provide an untrusted input that contains argument-separating delimiters, then the resulting command will have more arguments than intended by the developer, potentially changing the expected behavior of the command. Depending on the functionality supported by the extraneous arguments, this may have security-relevant consequences. The attacker could include arguments that allow unintended commands or code to be executed, allow sensitive data to be read or modified, or cause other unintended behavior.
Frequently Asked Questions
Not sure what something means? Check out our Security Glossary.
Want to get errata notifications? Sign up here.
