Story thread · 2 reports / 2 sources
Codex Sandbox Escapes Show Why Agent Guardrails Can’t Live Inside the Agent
devops.com · 2h
Two patched OpenAI Codex vulnerabilities, Heapjack and Overpatch, exposed how coding agents can escape sandboxes and reach developer systems without approval prompts.
First report: Researchers escape OpenAI Codex sandbox to run commands on host — bleepingcomputer.com, 1d
The conversation · 0
Sign in to join the conversation.
No comments yet — start the thread.