A Large-Scale Empirical Study of Secret Keys Leakage in Hugging Face Spaces
Published in IEEE/ACM 48th International Conference on Software Engineering (ICSE 2026), 2026
This paper presents the first large-scale and systematic empirical study to quantify the extent of secret key leakage in Hugging Face Spaces, identifying 9,149 vulnerable repositories and 11,557 unique leaked keys.
Recommended citation: Shaoxuan Yun, Yuchao Zhang, Zhikun Shi, Liu Wang, Yi Wang, and Yu Bai. (2026). "A Large-Scale Empirical Study of Secret Keys Leakage in Hugging Face Spaces." ICSE 2026.
Download Paper
