paper-with-me

홈 › Papers

Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents

2026-02-07 · Sai Puppala, Ismail Hossain, Md Jahangir Alam, Yoonpyo Lee, Jay Yoo, Tanzim Ahad, Syed Bahauddin Alam, Sajedul Talukder arxiv

Large language models are increasingly deployed as *deep agents* that plan, maintain persistent state, and invoke external tools, shifting safety failures from unsafe text to unsafe *trajectories*. We introduce AgentFence, an architecture-centric security evaluation that defines 14 trust-boundary attack classes spanning planning, memory, retrieval, tool use, and delegation, and detects failures via *trace-auditable conversation breaks* (unauthorized or unsafe tool use, wrong-principal actions, state/objective integrity violations, and attack-linked deviations). Holding the base model fixed, we evaluate eight agent archetypes under persistent multi-turn interaction and observe substantial architectural variation in mean security break rate (MSBR), ranging from $0.29 \pm 0.04$ (LangGraph) to $0.51 \pm 0.07$ (AutoGPT). The highest-risk classes are operational: Denial-of-Wallet ($0.62 \pm 0.08$), Authorization Confusion ($0.54 \pm 0.10$), Retrieval Poisoning ($0.47 \pm 0.09$), and Planning Manipulation ($0.44 \pm 0.11$), while prompt-centric classes remain below $0.20$ under standard settings. Breaks are dominated by boundary violations (SIV 31%, WPA 27%, UTI+UTA 24%, ATD 18%), and authorization confusion correlates with objective and tool hijacking ($ρ\approx 0.63$ and $ρ\approx 0.58$). AgentFence reframes agent security around what matters operationally: whether an agent stays within its goal and authority envelope over time.

📄 PDF Abstract BibTeX arXiv:2602.07652

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Improving Interoperability among Defence and National Security Ontologies: Analysis and Evaluation Tasks

2026-08-06 · Jonathon Dilworth, Pedro Giesteira Cotovio, David Herron, Paul Cripps 외 arxiv

The use of ontologies and knowledge graphs is becoming increasingly widespread in the defence and national security domain. Numerous ontologies have been developed through initiatives led by academia, industry, and gover…

Knowledge Graphs

Strategic commitments shape collective cybersecurity under AI inequality

2026-05-10 · Adeela Bashir, Zia Ush Shamszaman, Zhao Song, Matjaz Perc 외 arxiv

The growing integration of AI into cybersecurity is reshaping the balance between attackers and defenders. When access to advanced AI-enabled defence tools is uneven, resource-limited defenders may be unable to adopt eff…

Vulnerabilities of Connectionist AI Applications: Evaluation and Defence

2020-03-18 · Christian Berghoff, Matthias Neu, Arndt von Twickel

This article deals with the IT security of connectionist artificial intelligence (AI) applications, focusing on threats to integrity, one of the three IT security goals. Such threats are for instance most relevant in pro…

Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery

2026-09-09 · Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa, Sahil Agarwal 외 arxiv

Agentic systems are rapidly moving to production, where they read untrusted inputs, call tools with real permissions, and act autonomously, expanding the security surface beyond chat-only models. Yet standard evaluations…

Red Teaming

PathSeeker: Exploring LLM Security Vulnerabilities with a Reinforcement Learning-Based Jailbreak Approach

2024-09-21 · ZhiHao Lin, Wei Ma, Mingyi Zhou, Yanjie Zhao 외

In recent years, Large Language Models (LLMs) have gained widespread use, raising concerns about their security. Traditional jailbreak attacks, which often rely on the model internal information or have limitations when …

Multi-agent Reinforcement LearningSafety Alignment