analiza aiqbench
OpenAI Codex suffered from two critical sandbox escape vulnerabilities—Overpatch and Heapjack—that allowed malicious repositories to execute arbitrary commands on local systems without user consent. The Overpatch flaw exploited directory path manipulation in the apply_patch tool to breach file-write restrictions, while Heapjack compromised the node_repl environment within Codex Desk. These incidents underscore the urgent need for robust isolation mechanisms in AI agents that interact with untrusted codebases to prevent lateral movement from sandboxed environments to host operating systems.
OpenAI Codex Sandbox Flaws Let Malicious Repositories Execute Commands on Host Systems
OpenAI Codex recently faced two significant security vulnerabilities that allowed malicious repositories to potentially execute commands on a developer’s local system. These vulnerabilities, identified as Overpatch and Heapjack, were reported to OpenAI on August 12, 2026, and were addressed in under a week. The more critical issue, Heapjack, emerged when a developer accessed an attacker-controlled […] The post OpenAI Codex Sandbox Flaws Let Malicious Repositories Execute Commands on Host Systems appeared first on Cyber Security News.
OpenAI Codex recently faced two significant security vulnerabilities that allowed malicious repositories to potentially execute commands on a developer’s local system. These vulnerabilities, identified as Overpatch and Heapjack, were reported to OpenAI on August 12, 2026, and were addressed in under a week. The more critical issue, Heapjack, emerged when a developer accessed an attacker-controlled repository in Codex and queried the agent about its…
Extras afișat conform termenilor RSS ai publicației · 64 cuvinte · textul integral și figurile sunt pe site-ul publicației
Subiecte conexe
Friendly Fire: Hijacking Defensive Cyber AI Agents for Remote Code Execution
AI Now Institute · 8 iul. 2026
Weekly Cybersecurity Newsletter Bulletin – Cisco and Android 0-Day, BragJack Attack, Claude Opus 5 Used to Hack OpenAI, and 20+ Stories
Cybersecurity News · 21 sept. 2026
Critical Flaws in Anthropic, Google, and OpenAI’s Coding Agents Enable RCE and Supply Chain Attacks
Cybersecurity News · 6 aug. 2026
GitSpawn Flaws Let Malicious Repositories Execute Code in Claude Code, Codex, Cursor, and Grok
Cybersecurity News · 2 sept. 2026