Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
SkillSecurer: Detecting and Patching Prompt-Injection Vulnerabilities in AI Agent Skills
Donato Mecca, Alberto Verna, Youness Bouchari +2
Agent skills extend AI agents with reusable instructions, scripts, and configuration, but are also open to new attacks to influence an agent's decisions and actions. To address the…
cs.CR2026
Autonomous LLM Agents & CTFs: A Second Look
Youness Bouchari, Matteo Boffa, Marco Mellia +3
Large Language Model (LLM) agents are increasingly proposed to automate offensive security tasks, with recent studies reporting near human-level success rates in Capture-the-Flag (…