Hacker News AIT2·Industry
71 pts
Together AI disclosed a security incident in which an AI coding agent committed a credential to its public GitHub repository xorl, granting access to a private Hugging Face repo holding prompt samples from MiniMax M3, GLM-5.2 and Kimi K2.6 requests in June and July 2026. An unauthorized party downloaded the samples and searched them for further credentials. The incident highlights supply-chain risk from AI agents in development workflows; Together advises customers handling sensitive prompts to enable zero data retention.