Cover Image

Black-Box Skill Stealing Attack from Proprietary LLM Agents: An Empirical Study

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wang, Zihan, Zhang, Rui, Liu, Yu, Liu, Chi, Zhao, Qingchuan, Li, Hongwei, Xu, Guowen
Format:	Preprint
Published:	2026
Subjects:	Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MPMA: Preference Manipulation Attack Against Model Context Protocol
by: Wang, Zihan, et al.
Published: (2025)

BadLingual: A Novel Lingual-Backdoor Attack against Large Language Models
by: Wang, Zihan, et al.
Published: (2025)

BadTemplate: A Training-Free Backdoor Attack via Chat Template Against Large Language Models
by: Wang, Zihan, et al.
Published: (2026)

ConfGuard: A Simple and Effective Backdoor Detection for Large Language Models
by: Wang, Zihan, et al.
Published: (2025)

Combinational Backdoor Attack against Customized Text-to-Image Models
by: Jiang, Wenbo, et al.
Published: (2024)

Hidden Tail: Adversarial Image Causing Stealthy Resource Consumption in Vision-Language Models
by: Zhang, Rui, et al.
Published: (2025)

"Someone Hid It": Query-Agnostic Black-Box Attacks on LLM-Based Retrieval
by: Li, Jiate, et al.
Published: (2026)

Credential Leakage in LLM Agent Skills: A Large-Scale Empirical Study
by: Chen, Zhihao, et al.
Published: (2026)

Black-Box Guardrail Reverse-engineering Attack
by: Yao, Hongwei, et al.
Published: (2025)

PolyJailbreak: Cross-Modal Jailbreaking Attacks on Black-Box Multimodal LLMs
by: Wang, Xinkai, et al.
Published: (2025)

AdInject: Real-World Black-Box Attacks on Web Agents via Advertising Delivery
by: Wang, Haowei, et al.
Published: (2025)

The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
by: Zhang, Rui, et al.
Published: (2025)

A Hard-Label Black-Box Evasion Attack against ML-based Malicious Traffic Detection Systems
by: Liu, Zixuan, et al.
Published: (2025)

Efficient Data-Free Model Stealing with Label Diversity
by: Liu, Yiyong, et al.
Published: (2024)

Towards Effective Prompt Stealing Attack against Text-to-Image Diffusion Models
by: Zhao, Shiqian, et al.
Published: (2025)

OnePath: Efficient and Privacy-Preserving Decision Tree Inference in the Cloud
by: Yuan, Shuai, et al.
Published: (2024)

Trust Me, Import This: Dependency Steering Attacks via Malicious Agent Skills
by: Liu, Yiyong, et al.
Published: (2026)

InputSnatch: Stealing Input in LLM Services via Timing Side-Channel Attacks
by: Zheng, Xinyao, et al.
Published: (2024)

Assessing Risk of Stealing Proprietary Models for Medical Imaging Tasks
by: Raj, Ankita, et al.
Published: (2025)

GRID: Protecting Training Graph from Link Stealing Attacks on GNN Models
by: Lou, Jiadong, et al.
Published: (2025)

One Prompt to Verify Your Models: Black-Box Text-to-Image Models Verification via Non-Transferable Adversarial Attacks
by: Guo, Ji, et al.
Published: (2024)

SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement
by: Duan, Zenghao, et al.
Published: (2026)

Stealing Trust: Unraveling Blind Message Attacks in Web3 Authentication
by: Yan, Kailun, et al.
Published: (2024)

AttackPilot: Autonomous Inference Attacks Against ML Services With LLM-Based Agents
by: Wu, Yixin, et al.
Published: (2025)

Adversarial Agents: Black-Box Evasion Attacks with Reinforcement Learning
by: Domico, Kyle, et al.
Published: (2025)

When Skills Lie: Hidden-Comment Injection in LLM Agents
by: Wang, Qianli, et al.
Published: (2026)

Supply-Chain Poisoning Attacks Against LLM Coding Agent Skill Ecosystems
by: Qu, Yubin, et al.
Published: (2026)

BESA: Boosting Encoder Stealing Attack with Perturbation Recovery
by: Ren, Xuhao, et al.
Published: (2025)

Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale
by: Liu, Yi, et al.
Published: (2026)

Enhanced MLLM Black-Box Jailbreaking Attacks and Defenses
by: Zhong, Xingwei, et al.
Published: (2025)

Backdoor Attacks against Image-to-Image Networks
by: Jiang, Wenbo, et al.
Published: (2024)

Model Stealing Attack against Graph Classification with Authenticity, Uncertainty and Diversity
by: Zhu, Zhihao, et al.
Published: (2023)

Prompt Stealing Attacks Against Large Language Models
by: Sha, Zeyang, et al.
Published: (2024)

No Attack Required: Semantic Fuzzing for Specification Violations in Agent Skills
by: Li, Ying, et al.
Published: (2026)

AgentVigil: Generic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents
by: Wang, Zhun, et al.
Published: (2025)

Making Theft Useless: Adulteration-Based Protection of Proprietary Knowledge Graphs in GraphRAG Systems
by: Wang, Weijie, et al.
Published: (2026)

Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking
by: Zhang, Junke, et al.
Published: (2026)

A Model Stealing Attack Against Multi-Exit Networks
by: Pan, Li, et al.
Published: (2023)

Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents
by: Zhang, Hanrong, et al.
Published: (2024)

Can Drift-Adaptive Malware Detectors Be Made Robust? Attacks and Defenses Under White-Box and Black-Box Threats
by: Li, Adrian Shuai, et al.
Published: (2026)