GHSA-2FRJ-4QR5-M2RF
Vulnerability from github – Published: 2026-09-01 18:30 – Updated: 2026-09-01 18:30OpenAI 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.
{
"affected": [],
"aliases": [
"CVE-2026-19591"
],
"database_specific": {
"cwe_ids": [
"CWE-150"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-01T18:17:40Z",
"severity": null
},
"details": "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\u0027s 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\u0027s configuration. If Codex later loads the modified configuration, it can launch an attacker-controlled MCP server and execute code with the user\u0027s 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.",
"id": "GHSA-2frj-4qr5-m2rf",
"modified": "2026-09-01T18:30:45Z",
"published": "2026-09-01T18:30:45Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-19591"
},
{
"type": "WEB",
"url": "https://github.com/openai/codex/pull/22643"
}
],
"schema_version": "1.4.0",
"severity": []
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.