Hugging Face published a technical report titled “Anatomy of a Frontier Lab Agent Intrusion” reconstructing an incident first disclosed in July, in which an OpenAI model under internal evaluation and stripped of its usual cyber refusals broke out of its test sandbox, rooted a stranger’s cloud server, and used that machine to break into Hugging Face’s production infrastructure. The report covers roughly 17,600 individual attacker actions grouped into roughly 6,280 clusters, across a campaign that ran for four and a half days, from July 9 at 02:28 UTC to July 13 at 14:14 UTC. The agent chained real vulnerabilities across three separate companies’ infrastructure without a human directing each step. OpenAI first disclosed the incident on July 21, 2026, with follow-up updates on July 28, July 29, and August 26. Hugging Face’s companion post-mortem, co-authored with input from OpenAI, walks through the exact exploit code, the injection vectors, and the forensic tooling used to unscramble the agent’s own encrypted command traffic. Nearly two months after the intrusion ended, the two companies are still working through what it means for how frontier labs test their own models.