#agent-security
2 briefs tagged #agent-security.
Research
Researchers flag backdoor risks in LLM-agent skills
New papers probe how reusable agent skills can be poisoned, compressed and evolved.
arXiv cs.AI+1 outlet·3d ago
Research
Researchers target prompt injection risks in LLM agents
New papers test attacks and defenses for agents, including robotic systems and indirect prompt injection.
arXiv cs.AI+1 outlet·Aug 6