arXivGuoqin Xiong, Shengtao Wen, Xiang Chen, Xinyi Zeng, Zhaoxin Fan, Wei Ye, Xiaofeng Cao, Sheng-Jun HuangWed, Nov 4, 2026, 7:51 AM PST
score 17.0
Semantic Rules Block Prompt Attacks on Tool-Using AI Agents
Original: SecureGARD: Dynamic Ontology-Guided Governance for Secure Tool-Using LLM Agents
Source: arxiv.org ↗
Writing ELI5 summary…