Cover Image

Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool Feedback

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yan, Lecheng, Li, Ruizhe, Han, Xicheng, Li, Wenxi, Wang, Binwu, Wang, Longyue, Lyu, Chenyang, Chen, Guanhua
Format:	Preprint
Published:	2026
Subjects:	Cryptography and Security Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

AgentShield: Deception-based Compromise Detection for Tool-using LLM Agents
by: Rassul, Yassin H., et al.
Published: (2026)

Security Attacks on LLM-based Code Completion Tools
by: Cheng, Wen, et al.
Published: (2024)

MalTool: Malicious Tool Attacks on LLM Agents
by: Hu, Yuepeng, et al.
Published: (2026)

Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
by: Li, Xu, et al.
Published: (2026)

Measuring the Security of Mobile LLM Agents under Adversarial Prompts from Untrusted Third-Party Channels
by: Du, Chenghao, et al.
Published: (2025)

Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learning
by: An, Li, et al.
Published: (2025)

Imprompter: Tricking LLM Agents into Improper Tool Use
by: Fu, Xiaohan, et al.
Published: (2024)

AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender
by: Zhao, Weixiang, et al.
Published: (2025)

Les Dissonances: Cross-Tool Harvesting and Polluting in Pool-of-Tools Empowered LLM Agents
by: Li, Zichuan, et al.
Published: (2025)

Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools
by: Mohammadi, Bardia, et al.
Published: (2026)

Factor(T,U): Factored Cognition Strengthens Monitoring of Untrusted AI
by: Sandoval, Aaron, et al.
Published: (2025)

Overthinking Loops in Agents: A Structural Risk via MCP Tools
by: Lee, Yohan, et al.
Published: (2026)

VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commit
by: Lin, Junda, et al.
Published: (2026)

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models
by: Jiang, Peihai, et al.
Published: (2025)

InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents
by: Zhan, Qiusi, et al.
Published: (2024)

IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents
by: An, Hengyu, et al.
Published: (2025)

AgentTrust: Runtime Safety Evaluation and Interception for AI Agent Tool Use
by: Yang, Chenglin
Published: (2026)

ToolTweak: An Attack on Tool Selection in LLM-based Agents
by: Sneh, Jonathan, et al.
Published: (2025)

Causality Laundering: Denial-Feedback Leakage in Tool-Calling LLM Agents
by: Chinaei, Mohammad Hossein
Published: (2026)

STAC: When Innocent Tools Form Dangerous Chains to Jailbreak LLM Agents
by: Li, Jing-Jing, et al.
Published: (2025)

The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
by: Wang, Peiran, et al.
Published: (2026)

IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems
by: Wang, Liwen, et al.
Published: (2025)

SafeReview: Defending LLM-based Review Systems Against Adversarial Hidden Prompts
by: Xin, Yuan, et al.
Published: (2026)

TRUSTDESC: Preventing Tool Poisoning in LLM Applications via Trusted Description Generation
by: Ye, Hengkai, et al.
Published: (2026)

GuidedBench: Measuring and Mitigating the Evaluation Discrepancies of In-the-wild LLM Jailbreak Methods
by: Huang, Ruixuan, et al.
Published: (2025)

Iron Sharpens Iron: Defending Against Attacks in Machine-Generated Text Detection with Adversarial Training
by: Li, Yuanfan, et al.
Published: (2025)

Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
by: Qiao, Yuxuan, et al.
Published: (2025)

Preventing Jailbreak Prompts as Malicious Tools for Cybercriminals: A Cyber Defense Perspective
by: Tshimula, Jean Marie, et al.
Published: (2024)

A Broad Comparative Evaluation of Software Debloating Tools
by: Brown, Michael D., et al.
Published: (2023)

Prompt Injection Attack to Tool Selection in LLM Agents
by: Shi, Jiawen, et al.
Published: (2025)

Triad: Trusted Timestamps in Untrusted Environments
by: Fernandez, Gabriel P., et al.
Published: (2023)

From Defender to Devil? Unintended Risk Interactions Induced by LLM Defenses
by: Meng, Xiangtao, et al.
Published: (2025)

RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction
by: Jiang, Tanqiu, et al.
Published: (2024)

Attestable Builds: Compiling Verifiable Binaries on Untrusted Systems using Trusted Execution Environments
by: Hugenroth, Daniel, et al.
Published: (2025)

Watermark under Fire: A Robustness Evaluation of LLM Watermarking
by: Liang, Jiacheng, et al.
Published: (2024)

AgentVisor: Defending LLM Agents Against Prompt Injection via Semantic Virtualization
by: Ying, Zonghao, et al.
Published: (2026)

Watermarking LLM Agent Trajectories
by: Meng, Wenlong, et al.
Published: (2026)

MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents
by: Zhou, Zhenhong, et al.
Published: (2026)

Defensible Design for OpenClaw: Securing Autonomous Tool-Invoking Agents
by: Li, Zongwei, et al.
Published: (2026)

OpenClaw PRISM: A Zero-Fork, Defense-in-Depth Runtime Security Layer for Tool-Augmented LLM Agents
by: Li, Frank
Published: (2026)