Credential Leakage in LLM Agent Skills: A Large-Scale Empirical Study

Third-party skills extend LLM agents with powerful capabilities but often handle sensitive credentials in privileged environments, making leakage risks poorly understood. We present the first large-scale empirical study of this problem, analyzing 17,022 skills sampled from 170,226 on SkillsMP usi...

regexss

regexss Overly-greedy regex r...

Vulnerability Detection Pipeline (Beta)

Update October 22, 2020: The Vulnerability Detection Pipeline beta has been updated to include detections of all severities. It now gives visibility into upcoming and recently published detections with severity 3, 2 and 1 in addition to severity 5 and 4. The pipeline also supports a URL parameter...