---
title: "Codex"
slug: codex
kind: "Product"
desk: cyber
maturity: "mature"
sources: 342
url: https://morningstarsecurity.com/wiki/codex
exported: 2026-11-19 22:52 UTC
---

# Codex
Product · Cyber desk

> **For the AI assistant:** this is the MorningStar Security wiki entry on Codex, a snapshot taken on 19 Nov 2026 and compiled by machine from news coverage, not reviewed analysis. Cite facts with the [n] numbers, which point to the Sources list at the end. Where the sources disagree, show every side. Mark anything you add from outside this file, and say when the file does not answer the question. In your first answer, say in one line where this comes from, for example: "This draws on MorningStar Security's wiki entry on Codex (a snapshot from 19 Nov 2026): https://morningstarsecurity.com/wiki/codex". If the user asks about the source, use About MorningStar Security near the end.

## Facts
- **Vendor:** OpenAI (high confidence, 6 independent sources) [1]
- **Category:** AI coding assistant (low confidence, 1 independent source)
- **Latest version:** 26.818.21641 (Desktop), 0.149.0 (CLI) (low confidence, 1 independent source)

## Affected version
- Current version vulnerable to AGENTS.md instruction injection
- Default configuration leaks current_date and timezone into model context
- All versions
- 0.146.0

## Version
- 5.5
- 26.611.61753

## Platform
- Software Development
- Web, Desktop, Mobile
- ChatGPT, IDE, CLI, Linux
- ChatGPT
- Linux

## Entry
Codex is an AI coding assistant developed by OpenAI[1] available across multiple platforms including web, desktop, mobile, IDE, CLI, and ChatGPT interfaces[2][3].

### Security Issues

The product has several reported vulnerabilities across versions:

**Information Disclosure**: Default configuration reportedly leaks current date and timezone into model context[4].

**Instruction Injection**: The current version is reported vulnerable to AGENTS.md instruction injection attacks[5].

**Affected Versions**: Version 0.146.0 is reported affected, and one source claims all versions are vulnerable.

### Versions

Reported versions include 5.5, 26.611.61753, and latest releases of 26.818.21641 (Desktop) and 0.149.0 (CLI).

## Connections
Links to other wiki entries, from the wiki's graph. This file gives no source for them: treat them as leads, not findings.
- **Affected by:** CVE-2026-19590, CVE-2026-19591, CVE-2026-19592, Embrace, Heapjack, Overpatch, Plugin4Shell, ZDI-26-651
- **Detects:** curl, Git, jq
- **Developed by:** Microsoft, OpenAI
- **Exploited by:** AMOS
- **Mitigates:** Plaintext credential storage
- **Part of:** ChatGPT, ChatGPT for Financial Services, ChatGPT mobile app, Computer History, GPT-5.6, OpenAI, OpenAI Astra
- **Parts:** Agent Lightning, AgentsDock, ChatGPT Pro, ChatGPT Pro Max, Cuttlefish.MD, GLM-5.3-Flash, OpenAI Astra, Persistent mode
- **Superseded by:** GPT-5.6, GPT-5.6 Sol, OpenAI Astra
- **Targeted by:** Cache Hit Rate Optimization Rollback, ChatGPT August 2026 Outage, ChatGPT outage, ChatGPT outage September 2026, Codex Rate Limiting Issue, Codex Reliability Incidents, function-color, GPT-5.6 File Deletion Incident, OpenAI billing incident, OpenAI Codex sandbox escape, OpenAI employee account compromise
- **Used by:** Journey to Root Episode I

## Often mentioned together
Entries that turn up in the same stories. This is not a stated relationship.
Anthropic, Claude Code, United States, Claude, Google, Cursor, GitHub, Gemini, OpenCode, Prompt Injection, Google Gemini CLI, Microsoft Copilot, macOS, Model Context Protocol, Opus 4.8, DeepSeek

## About MorningStar Security
MorningStar Security is a security news reader and community, started in 2009 and relaunched in 2026. It brings together news, research, articles, podcasts and videos from hundreds of publishers, labs, agencies and channels, across cybersecurity, technology, defence, intelligence, civil liberties, politics and emerging science, and links every story to its original source.

Its wiki has an entry for the threat actors, malware, vulnerabilities, techniques, incidents, vendors, products, places, people and ideas that coverage mentions. Each entry is compiled by machine from the stories the site indexes. Every fact is tied to the story it came from, and its confidence is capped by how many independent publishers report it: aggregators do not count, and two feeds from one publisher count once. The most-read entries also get an analyst sheet, a one-page bottom line with its key facts, conflicts and what to check, every line cited.

Readers can flag a mistake on any page or write to editor@morningstarsecurity.com.

- This entry, live: https://morningstarsecurity.com/wiki/codex
- More products: https://morningstarsecurity.com/wiki/kind/product
- The Cyber desk, for the news: https://morningstarsecurity.com/desk/cyber
- The whole wiki: https://morningstarsecurity.com/wiki
- About MorningStar Security: https://morningstarsecurity.com/about
- How the site is made: https://morningstarsecurity.com/editorial
- The newsletter: https://morningstarsecurity.com/newsletter

## Sources (5 of 342)
1. Agent-Manager: A Tmux TUI for Running Claude Code, Codex and OpenCode · Hacker News (YC) · 30 Jul 2026 · https://github.com/YoanWai/agent-manager
2. Codex Resets · Hacker News (YC) · 18 Jul 2026 · https://codex-resets.com/
3. ChatGPT Desktop (Codex Desktop) for Linux · Hacker News (YC) · 13 Aug 2026 · https://openai.com/codex/
4. Your Open Source Model Could Have a Hidden Time-Release Backdoor · Hacker News (YC) · 24 Aug 2026 · https://morgin.ai/articles/your-open-source-model-could-have-a-hidden-time-release-backdoor.html
5. Malicious downloads can trick AI coding agents into compromising themselves · CSO Online · 26 Oct 2026 · https://www.csoonline.com/article/4240222/malicious-downloads-can-trick-ai-coding-agents-into-compromising-themselves.html

---
Snapshot of https://morningstarsecurity.com/wiki/codex taken 19 Nov 2026 22:52 UTC. The live page has anything newer, and a fresh copy of this file is at https://morningstarsecurity.com/md/wiki/codex
