Trusted Credentials, Untrusted Behavior: Benchmarking LLM-Agent Security in High-Performance Computing

Large language model LLM agents are starting to take on routine work in high-performance computing HPC, including monitoring Slurm jobs, diagnosing failed builds, inspecting simulation output, and coordinating scientific workflows. To do this work, an agent commonly acts under its user's...

EUVD-2024-42809

Malicious code in bioql PyPI...

Linux Distros Unpatched Vulnerability : CVE-2019-9892

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - An issue was discovered in Open Ticket Request System OTRS 5.x through 5.0.34, 6.x through 6.0.17, and 7.x through 7.0.6. An attacker who is logged into OTRS as...

Design/Logic Flaw

An authenticated and authorized agent user could potentially gain administrative access via an SQLi vulnerability to Capsule8 Console between versions 4.6.0 and 4.9.1...