11 papers
AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent
Weidi Luo, Qiming Zhang, Yihao Quan +5
Coding agents based on large language models (LLMs) demonstrate remarkable autonomous capabilities, but they also introduce significant safety and misuse risks during multi-turn in…
AutoDojo: Adaptive Black-Box Attacks Reveal the Limits of IPI Defenses and Task-Specification Effects in LLM Agents
Xinhang Ma, Taoran Li, Chaowei Xiao +3
Indirect prompt injection (IPI) is a major security threat to LLM-powered agents. Thus, a growing body of work have proposed a variety of defensive approaches against IPI. These ca…
FORTIS: Benchmarking Over-Privilege in Agent Skills
Shawn Li, Chenxiao Yu, Han Wang +8
Large language model agents increasingly operate through an intermediate skill layer that mediates between user intent and concrete task execution. This layer is widely treated as…
Runtime Skill Audit: Targeted Runtime Probing for Agent Skill Security
Tu Lan, Chaowei Xiao
Agent skills let LLM agents reuse instructions, resources, tools, and workflows, but they also create a new place for malicious behavior to hide. A skill may look benign in its doc…
AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?
Hao Li, Ruoyao Wen, Shanghao Shi +3
AI agents that autonomously interact with external tools and environments have shown great promise across real-world applications. However, their reliance on external data exposes…
ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems
Zhuowen Yuan, Zhaorun Chen, Zhen Xiang +5
Existing research on LLM agent security mainly focuses on prompt injection and unsafe input/output behaviors. However, as agents increasingly rely on third-party tools and MCP serv…