182 Credentials Hid in 'Encrypted' Reasoning. Go Rotate.
Researchers decoded 315,320 encrypted reasoning blocks from 6,708 agent trajectories published to GitHub and Hugging Face, recovering 182 credentials and 367 PII artifacts. 64 of the 704 artifacts recovered from genuine user sessions never appeared in the visible chat history, which means transcript review provably misses some of what you have already published.
August 12, 2026 · 15 min read