
jataayu
LLM-backed AI agent security — inbound injection detection + outbound privacy protection
ai-securityauthentication-authorizationdata-exfiltration+6

LLM-backed AI agent security — inbound injection detection + outbound privacy protection
CVE-2026-32662: Active Debug Code in Production — Gardyn Home Kit (ICSA-26-055-03)

OpenGraph collector for BloodHound that maps attack paths from DevOps to MLOps infrastructure, collecting CI/CD pipeline, service principal, and ML…