# CVE-2026-110930: Codex Desktop for Windows code execution via unquoted path in file explorer

Source: https://junglewise.ai/threats/cve-2026-110930-codex-desktop-for-windows-does-not-correctly-check-quotation  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-11-05. Updated: 2026-11-05  
License: CC BY 4.0 with a link back

CVE-2026-110930 · Severity: info · Published 2026-11-05

Codex Desktop for Windows is a code editor that integrates with ChatGPT. An attacker can craft a malicious library folder that, when a user opens it in ChatGPT through Windows Explorer, injects command-line arguments that allow arbitrary code execution with the user's permissions before authentication or security checks occur.

Technologies: [OpenAI Codex Desktop](https://junglewise.ai/threats/technologies/codex-desktop). Vendors: [OpenAI](https://junglewise.ai/threats/vendors/openai).

## Quick answers

**Is CVE-2026-110930 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-110930 being exploited in the wild.

**How severe is CVE-2026-110930?** Informational severity, published 5 November 2026.

**Which products and versions are affected?** OpenAI Codex Desktop <UNKNOWN>.

**Is there a fix for CVE-2026-110930?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-110930 attacked?** It is attacked with local access, in a web application.

**Am I affected by CVE-2026-110930?** Check whether you run OpenAI Codex Desktop at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

The vulnerability is a command-line argument injection via improper quotation mark handling in project paths. When a user selects "Open project in ChatGPT" from Windows Explorer on an attacker-controlled child item, the application fails to properly quote the path parameter, allowing injection of additional startup arguments. An unauthenticated attacker can execute arbitrary code with the current Windows user's privileges before sign-in or project trust approval is completed.

## Affected products

- OpenAI Codex Desktop <UNKNOWN>

## Timeline

- 2026-11-05: disclosed

## References

- [https://openai.com/security/advisories/cve-2026-110930/](https://openai.com/security/advisories/cve-2026-110930/)

## Related threats

- [OpenAI Codex Desktop arbitrary code execution via Git config](https://junglewise.ai/threats/cve-2026-19593-openai-codex-desktop-arbitrary-code-execution-via-git-config)
- [OpenAI Codex CLI and Desktop Git fsmonitor code execution](https://junglewise.ai/threats/cve-2026-19592-openai-codex-cli-and-desktop-git-fsmonitor-code-execution)
- [OpenAI Codex CLI and Desktop PowerShell command-safety bypass](https://junglewise.ai/threats/cve-2026-19591-openai-codex-cli-and-desktop-powershell-command-safety-bypass)
- [OpenAI Codex Desktop arbitrary code execution via Git hooks](https://junglewise.ai/threats/cve-2026-19590-openai-codex-desktop-arbitrary-code-execution-via-git-hooks)