SecureGARD: Dynamic Ontology-Guided Governance for Secure Tool-Using LLM Agents

Summary

SecureGARD is a dynamic ontology-guided governance framework for tool-using LLM agents. It maps heterogeneous tool calls to a shared semantic representation and uses an executable ontology with episode-level task state to issue ready, waiting, or forbidden decisions. Evaluated on AgentDojo, ASB, and AgentDyn under benign and indirect prompt-injection settings, it reports a 0% attack success rate across all three benchmarks while improving task utility under attack.

Related items

• MediumCVE-2026-122235: Jan workspace guard bypass allows file writes outside workspace via symlinksSimilar attack · NVD/CVE Database
• InfoTop agent skill security resources — November 2026Similar attack · Adversa AI Blog
• InfoSecuring Web Agents with the Compile-Then-Execute ParadigmSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoWhen File Ingestion Misleads Language Models: Measuring and Mitigating Divergences in Office DocumentsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoHow to Turn AI Red Team Findings Into Runtime ControlsSimilar attack · Straiker Blog