Imagen de Portada

Uncovering Security Threats and Architecting Defenses in Autonomous Agents: A Case Study of OpenClaw

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Ying, Zonghao, Yang, Xiao, Wu, Siyang, Song, Yumeng, Qu, Yang, Li, Hainan, Li, Tianlin, Wang, Jiakai, Liu, Aishan, Liu, Xianglong
Formato:	Preprint
Publicado:	2026
Materias:	Cryptography and Security
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

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

Taming OpenClaw: Security Analysis and Mitigation of Autonomous LLM Agent Threats
por: Deng, Xinhao, et al.
Publicado: (2026)

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw
por: Yao, Hongwei, et al.
Publicado: (2026)

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

Don't Let the Claw Grip Your Hand: A Security Analysis and Defense Framework for OpenClaw
por: Shan, Zhengyang, et al.
Publicado: (2026)

Security, Privacy, and Ethical Risks in OpenClaw
por: Jin, Yutong, et al.
Publicado: (2026)

Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study
por: Xu, Luyao, et al.
Publicado: (2026)

A Security Analysis of the OpenClaw AI Agent Framework
por: Suwansathit, Surada, et al.
Publicado: (2026)

A Systematic Security Evaluation of OpenClaw and Its Variants
por: Wang, Yuhang, et al.
Publicado: (2026)

ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers
por: Liu, Songyang, et al.
Publicado: (2026)

Agent Privilege Separation in OpenClaw: A Structural Defense Against Prompt Injection
por: Cheng, Darren, et al.
Publicado: (2026)

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

Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks
por: Ying, Zonghao, et al.
Publicado: (2024)

A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)
por: Chen, Tianyu, et al.
Publicado: (2026)

Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw
por: Wang, Zijun, et al.
Publicado: (2026)

AGENTSAFE: Benchmarking the Safety of Embodied Agents on Hazardous Instructions
por: Ying, Zonghao, et al.
Publicado: (2025)

ClawTrap: A MITM-Based Red-Teaming Framework for Real-World OpenClaw Security Evaluation
por: Zhao, Haochen, et al.
Publicado: (2026)

Evolving Deception: When Agents Evolve, Deception Wins
por: Ying, Zonghao, et al.
Publicado: (2026)

Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance
por: Liu, Fazhong, et al.
Publicado: (2026)

Clawdrain: Exploiting Tool-Calling Chains for Stealthy Token Exhaustion in OpenClaw Agents
por: Dong, Ben, et al.
Publicado: (2026)

MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study
por: Van hamme, Tim, et al.
Publicado: (2026)

Foundations for Agentic AI Investigations from the Forensic Analysis of OpenClaw
por: Gruber, Jan, et al.
Publicado: (2026)

SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents
por: Ying, Zonghao, et al.
Publicado: (2025)

Compromising Embodied Agents with Contextual Backdoor Attacks
por: Liu, Aishan, et al.
Publicado: (2024)

SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models
por: Ying, Zonghao, et al.
Publicado: (2024)

Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt
por: Ying, Zonghao, et al.
Publicado: (2024)

SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents
por: Cheng, Hao, et al.
Publicado: (2026)

GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic
por: Zhang, Tianyuan, et al.
Publicado: (2026)

Execution Is the New Attack Surface: Survivability-Aware Agentic Crypto Trading with OpenClaw-Style Local Executors
por: Borjigin, Ailiya, et al.
Publicado: (2026)

SPARK: Jailbreaking T2V Models by Synergistically Prompting Auditory and Recontextualized Knowledge
por: Ying, Zonghao, et al.
Publicado: (2025)

ClawLess: A Security Model of AI Agents
por: Lu, Hongyi, et al.
Publicado: (2026)

Deep Learning Model Security: Threats and Defenses
por: Wang, Tianyang, et al.
Publicado: (2024)

Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings
por: Ying, Zonghao, et al.
Publicado: (2025)

QuantClaw: Precision Where It Matters for OpenClaw
por: Zhang, Manyi, et al.
Publicado: (2026)

Towards Secure Retrieval-Augmented Generation: A Comprehensive Review of Threats, Defenses and Benchmarks
por: Mu, Yanming, et al.
Publicado: (2026)

Ambush from All Sides: Understanding Security Threats in Open-Source Software CI/CD Pipelines
por: Pan, Ziyue, et al.
Publicado: (2024)

Architecting Secure AI Agents: Perspectives on System-Level Defenses Against Indirect Prompt Injection Attacks
por: Xiang, Chong, et al.
Publicado: (2026)

Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models
por: Ying, Zonghao, et al.
Publicado: (2025)

OpenClaw-RL: Train Any Agent Simply by Talking
por: Wang, Yinjie, et al.
Publicado: (2026)

Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions
por: Li, Wenjuan, et al.
Publicado: (2026)