alixiacf/hpim-training-lab

PoC

Trained to Escalate: Forensic Analysis and Local Replication of RLHF-Induced Privilege Escalation in AI Agents (CVE-2026-65616)

view exploit source poc_in_githubauthor alixiacfpublished 2026-09-19catalog poc_in_github

exploited CVEs (1)

Known CVEs related to this entry.

ID	severity	EPSS	exploitable	published
CVE-2026-65616US	—	0.37% (31st)	Exploit	2026-07-27
Incorrect authorization validation in refresh token signature allows non-admin users to obtain a signed JFrog administrator token.

scry

Working exploits like this target internet-facing systems. See what is exposed on your perimeter with a free passive scan.