CVE-2026-19591

Description

OpenAI Codex CLI for Windows, macOS, and Linux and Codex Desktop for Windows and macOS misclassified certain PowerShell commands as safe because their command-safety parser interpreted PowerShell's stop-parsing token (--%) differently than PowerShell itself. If a user opens an attacker-prepared repository and Codex follows its instructions, Codex can run a file-writing Git command without requesting user approval. On macOS and Linux, exploitation additionally requires separately installed PowerShell Core (pwsh) to be invoked. If filesystem protections permit the write, the command can modify Codex's configuration. If Codex later loads the modified configuration, it can launch an attacker-controlled MCP server and execute code with the user's privileges, allowing it to read, change, or delete files accessible to that account. The approval bypass does not disable filesystem sandboxing; the default filesystem sandbox on macOS and Linux can prevent writes outside permitted locations.

Is your site exposed to CVE-2026-19591?

Run a free security scan — no signup, results in seconds.

Weakness Type (CWE)

CWE-150 CWE-150

References

Other References

https://github.com/openai/codex/pull/22643

Frequently Asked Questions

What is CVE-2026-19591? +

OpenAI Codex CLI for Windows, macOS, and Linux and Codex Desktop for Windows and macOS misclassified certain PowerShell commands as safe because their command-safety parser interpreted PowerShell's stop-parsing token (--%) differently than PowerShell itself. If a user opens an attacker-prepared repository and Codex follows its instructions, Codex can run a file-writing Git command without requesting user approval. On macOS and Linux, exploitation additionally requires separately installed PowerShell Core (pwsh) to be invoked. If filesystem protections permit the write, the command can modify Codex's configuration. If Codex later loads the modified configuration, it can launch an attacker-controlled MCP server and execute code with the user's privileges, allowing it to read, change, or delete files accessible to that account. The approval bypass does not disable filesystem sandboxing; the default filesystem sandbox on macOS and Linux can prevent writes outside permitted locations.

How do I check if I'm vulnerable to CVE-2026-19591? +

You can use Secably's free Website Scanner to check your website for known vulnerabilities. For infrastructure scanning, use the Port Scanner to identify exposed services that may be affected. Check the vendor advisories linked above for specific patch and version information.

Related Vulnerabilities

CVE-2026-6019

http.cookies.Morsel.js_output() returns an inline <script> snippet and only escapes " for JavaScript string context. It does not neutralize the HTML …

CVE-2025-58160

tracing is a framework for instrumenting Rust programs to collect structured, event-based diagnostic information. Prior to version 0.3.20, tracing-subscriber was …

CVE-2025-55193

Active Record connects classes to relational database tables. Prior to versions 7.1.5.2, 7.2.2.2, and 8.0.2.1, the ID passed to find …

CVE-2025-47284 9.9

Gardener implements the automated management and operation of Kubernetes clusters as a service. A security vulnerability was discovered in the …

CVE-2026-11362 9.8

DataDog::DogStatsd versions through 0.07 for Perl allow metric injections from event tags. DataDog::DogStatsd does not properly sanitise input, allowing metric …

CVE-2025-25286 9.8

Crayfish is a collection of Islandora 8 microservices, one of which, Homarus, provides FFmpeg as a microservice. Prior to Crayfish …