LLM Agent Security

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

29 papers in the last four weeks, up 81% on the four weeks before. 0.3% of all new papers.

Jul 13Week of Sep 28

Latest papers 234

All topics

StealthBench: Measuring Operational Stealth in Autonomous Offensive-Security Agents
Jul 28, 2026Ads Dawson, Adrian WoodAI Agent Security LLM Agent Security
Security Researcher, USA
Distributing Security Controls Through Harness Engineering
Jul 28, 2026William Robert GoreAI Agent Security LLM Agent Security
Georgia Institute of Technology
Agent Skills Matter: Inferring Proprietary Skills from Execution Trajectories
Jul 28, 2026Jianing Geng, Ruiqi He, Zekun Fei +6LLM Agent Security
Nankai University · Alibaba Group · East China University of Science and Technology +2
Cyber-Capable AI Agents: Vulnerabilities, Evaluation Containment, and Defensive Response
Jul 28, 2026Abu Bakar SiddikAI Agent Evaluation Cybersecurity
Department of Computer Science and Engineering, Rajshahi University of Engineering & Technology, Rajshahi 6204, Bangladesh
Explanation-Bound Tool Execution for AI Agents: Server-Verified Action Claims Without Trusting Model Rationales
Jul 28, 2026Genliang Zhu, Chu WangLLM Agent Security LLM Agent Verification
Accentrust · Georgia Institute of Technology · University of Illinois Urbana-Champaign
Hybrid Analysis for Secure MCP Tool Use in LLM Agents
Jul 28, 2026Ping He, Yuexiang Xie, Yaliang Li +1MCP Security LLM Agent Security
Zhejiang University · Alibaba Group
Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents
Jul 27, 2026Arseny Kravchenko, Vadim Liventsev, Innokentii Konstantinov +2LLM Agent Security Prompt Injection Attacks on AI Agents
Archestra AI
Agentic Cloud Decoys: A Deception-Driven Framework for Autonomous Intrusion Investigation
Jul 27, 2026Mohan Manivannan, Dalal AlharthiAutonomous Cyber Defense Cybersecurity
College of Information Science University of Arizona Tucson, AZ, USA
Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
Jul 24, 2026Halil Burak NoyanLLM Agent Security AI Agent Security Benchmarks
Independent Researcher, London, United Kingdom.
ToolGuardian: Declarative Security for AI Agent-Tool Interactions
Jul 23, 2026Arun Ravindran, Saurabh DeochakeAI Agent Security LLM Agent Security
Department of Electrical and Computer Engineering University of North Carolina at Charlotte Charlotte, NC · SentinelOne Mountain View, CA
Twin Agent: Context Residual Compression for Privilege Separated Agents
Jul 21, 2026Zhanhao Hu, Dennis Jacob, Xiao Huang +3AI Agent Security LLM Agent Security
University of California, Berkeley · University of Chicago · University of Illinois, Urbana-Champaign
They'll Verify. They Just Won't Act. How Authority Framing and Laundered Code Turn a Trusted Agentic CI/CD Pipeline Into an Attack Surface
Jul 21, 2026Yohann SidotLLM Agent Security Indirect Prompt Injection
Senthex Research
Data Leakage Prevention in Agentic Applications via Preemptive Hardening
Jul 21, 2026Akansha Shukla, Emily Bellov, Parth Atulbhai Gandhi +2AI Agent Security LLM Agent Security
Faculty of Computer and Information Science Ben-Gurion University of the Negev Beer-Sheva, Israel
Broken Gates: Re-evaluating Web Bot Defenses in the Age of LLM Agents
Jul 21, 2026Behzad Ousat, Nikita Turkmen, Lalchandra Rampersaud +2LLM Agent Security Web Agents
Florida International University
Trusted Credentials, Untrusted Behavior: Benchmarking LLM-Agent Security in High-Performance Computing
Jul 20, 2026Jie LiLLM Agent Security AI Agent Security Benchmarks
Texas Tech University
ChannelGuard: Safe Models Do Not Compose into Safe Multi-Agent Systems
Jul 20, 2026Elias Hossain, Md Mehedi Hasan Nipu, Fatema Tuj Johora Faria +2LLM Agent Security Prompt Injection Attacks on AI Agents
University of Central Florida · 2North South University · 3Ahsanullah University of Science and Technology +1
SlotGuard: Stop Oversharing Private Local Context in LLM Agent Transcri
Jul 19, 2026Haocheng Xia, Yongjoo ParkLLM Agent Security Privacy Leakage in Language Models
Siebel School of Computing and Data Science, University of Illinois Urbana-Champaign.
MemPoison: Uncovering Persistent Memory Threats and Structural Blind Spots in LLM Agents
Jul 16, 2026Jifeng Gao, Kang Xia, Yi Zhang +5LLM Agent Memory LLM Agent Security
State Key Laboratory for Novel Software Technology, Nanjing University, China · NARI Group Corporation/State Grid Electric Power Research Institute, China
Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions
Jul 14, 2026Huihao Jing, Wenbin Hu, Shaojin Chen +10LLM Agent Security LLM Agent Safety
HKUST · NYU · SWUPL
Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming
Jul 13, 2026Xutao Mao, Xiang Zheng, Cong WangLLM Agent Security Automated Red Teaming
City University of Hong Kong
Temporary Authority, Permanent Effects: Commit-Time Authorization for LLM Agents
Jul 11, 2026Igor Santos-GrueiroSoftware Security LLM Agent Security
International University of La Rioja
VEXAIoT: Autonomous IoT Vulnerability EXploitation using AI Agents
Jul 10, 2026Katherine Swinea, Kshitiz Aryal, Lopamudra Praharaj +1Cybersecurity LLM Agent Security
Department of Computer Science, Tennessee Tech University, TN, USA. · School of Interdisciplinary Informatics, University of Nebraska Omaha, NE, USA. · Dept. of Mathematics & Computer Science, University of North Carolina at Pembroke, NC, USA.
TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories
Jul 9, 2026Zheng Gao, Xiaoyu Li, Xiaoyan Feng +5LLM Agent Security LLM Watermarking
University of New South Wales · 2Griffith University · 3CSIRO’s Data61
Out of Sight: Compression-Aware Content Protection against Agentic Crawlers
Jul 9, 2026Xuefei WangLLM Agent Security Black-Box Adversarial Attacks
Beihang University
Prismata: Confining Cross-Site Prompt Injection in Web Agents
Jul 9, 2026Corban Villa, Alp Eren Ozdarendeli, Sijun Tan +1LLM Agent Security Information Flow Control
UC Berkeley
When Agents Go Rogue: Activation-Based Detection of Malicious Behaviors in Multi-Agent Systems
Jul 7, 2026Haowen Xu, Xue Tan, Lei Ma +6LLM Agent Security AI Agent Monitoring
Department of Computer Science, Worcester Polytechnic Institute, MA, USA · School of Computer Science, Fudan University, Shanghai, China · Institute of Big Data, Fudan University, Shanghai, China +1
When Agents Remember Too Much: Memory Poisoning Attacks on Large Language Model Agents
Jul 6, 2026George Torres, Sharad Shrestha, Satyajayant MisraAI Agent Security LLM Agent Security
‡∗§Department of Computer Science, §Klipsch School of Electrical and Computer Engineering New Mexico State University, Las Cruces, New Mexico, USA · §Klipsch School of Electrical and Computer Engineering New Mexico State University, Las Cruces, New Mexico, USA
aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents
Jul 6, 2026Sai Varun KodathalaAI Agent Security LLM Agent Security
Research & Development SportsVision AI Minnetonka, MN
Biological Motifs for Agentic Control
Jul 5, 2026Bogdan BanuMulti-Agent LLM Systems LLM Agent Security
DualView: Preventing Indirect Prompt Injection in Personal AI Agents
Jul 4, 2026Juhee Kim, Woohyuk Choi, Taehyun Kang +2LLM Agent Security Indirect Prompt Injection
Seoul National University

Jul 28, 2026, cs.CRJ/K move · Enter open · S save

StealthBench: Measuring Operational Stealth in Autonomous Offensive-Security Agents

Ads Dawson, Adrian Wood

Security Researcher, USA

AI Agent Security LLM Agent Security AI Agent Security Benchmarks Automated Penetration Testing

Open paper AskPDF arXiv

Stealth, the discipline of achieving an objective without revealing your presence, capabilities, or collected intelligence, is what separates sophisticated operators from detectable ones. Elite security researchers and advanced persistent threats achieve their objectives unnoticed; autonomous agents increasingly inherit the same offensive tasks, but do they inherit the tradecraft? We introduce StealthBench,a benchmark that measures operational stealth in autonomous offensive-security agents across six operational security (OPSEC) dimensions. We extract 11 hand-verified OPSEC incidents from real bug-bounty and red-team trajectories, expanded into 14 dockerized task scenarios, where agents, despite finding real vulnerabilities, committed stealth failures inconsistent with standard operational tradecraft: embedding credentials in public uploads, deleting production resources to prove access, force-adding uninvolved users to demonstrate a race condition. We evaluate agent trajectories using a 3-model large language model (LLM) judge panel with majority-vote aggregation, measuring safe success rate (solved and stealthy), Stealth@Solve (tradecraft quality among successful solves), and reckless solve rate (solved but cover blown). Our results show that no model exceeds 54% safe success rate (the compound metric requiring both task completion and stealth), confirming that OPSEC failures are systematic across model families. We release StealthBench as a public benchmark to support both the development of stealth-aware agents and automated OPSEC monitoring for autonomous offensive-security deployments. The interactive leaderboard, evaluation harness, and dataset are available at https://stealthbench.com.

Momentum

Related topics

Latest papers 234