Copertina

Bypassing AI Control Protocols via Agent-as-a-Proxy Attacks

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Isbarov, Jafar, Kantarcioglu, Murat
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Cryptography and Security Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Optimal Transport-Guided Adversarial Attacks on Graph Neural Network-Based Bot Detection
di: Mukherjee, Kunal, et al.
Pubblicazione: (2026)

Agent Control Protocol: Admission Control for Agent Actions
di: Fernandez, Marcelo
Pubblicazione: (2026)

Adaptive Attacks on Trusted Monitors Subvert AI Control Protocols
di: Terekhov, Mikhail, et al.
Pubblicazione: (2025)

Attention Masks Help Adversarial Attacks to Bypass Safety Detectors
di: Shi, Yunfan
Pubblicazione: (2024)

MCP Bridge: A Lightweight, LLM-Agnostic RESTful Proxy for Model Context Protocol Servers
di: Ahmadi, Arash, et al.
Pubblicazione: (2025)

IPI-proxy: An Intercepting Proxy for Red-Teaming Web-Browsing AI Agents Against Indirect Prompt Injection
di: Chia-Pei, et al.
Pubblicazione: (2026)

MCP Security Bench (MSB): Benchmarking Attacks Against Model Context Protocol in LLM Agents
di: Zhang, Dongsen, et al.
Pubblicazione: (2025)

Automatically Attacking Software Reverse Engineering AI Agents
di: Crawford, Brian, et al.
Pubblicazione: (2026)

Securing AI Agents Against Prompt Injection Attacks
di: Ramakrishnan, Badrinath, et al.
Pubblicazione: (2025)

Emerging Cyber Attack Risks of Medical AI Agents
di: Qiu, Jianing, et al.
Pubblicazione: (2025)

Agentic JWT: A Secure Delegation Protocol for Autonomous AI Agents
di: Goswami, Abhishek
Pubblicazione: (2025)

Prompts Don't Protect: Architectural Enforcement via MCP Proxy for LLM Tool Access Control
di: Uppala, Rohith
Pubblicazione: (2026)

ATAG: AI-Agent Application Threat Assessment with Attack Graphs
di: Gandhi, Parth Atulbhai, et al.
Pubblicazione: (2025)

OpenPort Protocol: A Security Governance Specification for AI Agent Tool Access
di: Zhu, Genliang, et al.
Pubblicazione: (2026)

AESP: A Human-Sovereign Economic Protocol for AI Agents with Privacy-Preserving Settlement
di: Wang, Jian Sheng
Pubblicazione: (2026)

From Prompt Injections to Protocol Exploits: Threats in LLM-Powered AI Agents Workflows
di: Ferrag, Mohamed Amine, et al.
Pubblicazione: (2025)

Stealthy Poisoning Attacks Bypass Defenses in Regression Settings
di: Carnerero-Cano, Javier, et al.
Pubblicazione: (2026)

CheatAgent: Attacking LLM-Empowered Recommender Systems via LLM Agent
di: Ning, Liang-bo, et al.
Pubblicazione: (2025)

MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents
di: Zhu, Kaijie, et al.
Pubblicazione: (2025)

AutoBackdoor: Automating Backdoor Attacks via LLM Agents
di: Li, Yige, et al.
Pubblicazione: (2025)

Securing AI Agents with Information-Flow Control
di: Costa, Manuel, et al.
Pubblicazione: (2025)

Progent: Securing AI Agents with Privilege Control
di: Shi, Tianneng, et al.
Pubblicazione: (2025)

Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks
di: Saha, Shoumik, et al.
Pubblicazione: (2025)

LLM-driven Provenance Forensics for Threat Investigation and Detection
di: Mukherjee, Kunal, et al.
Pubblicazione: (2025)

Whispers of Wealth: Red-Teaming Google's Agent Payments Protocol via Prompt Injection
di: Debi, Tanusree, et al.
Pubblicazione: (2026)

Under the Hood of SKILL.md: Semantic Supply-chain Attacks on AI Agent Skill Registry
di: Saha, Shoumik, et al.
Pubblicazione: (2026)

Fast Proxies for LLM Robustness Evaluation
di: Beyer, Tim, et al.
Pubblicazione: (2025)

BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning
di: Tie, Guiyao, et al.
Pubblicazione: (2026)

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

Security of Internet of Agents: Attacks and Countermeasures
di: Wang, Yuntao, et al.
Pubblicazione: (2025)

CompressionAttack: Exploiting Prompt Compression as a New Attack Surface in LLM-Powered Agents
di: Liu, Zesen, et al.
Pubblicazione: (2025)

ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying
di: Lyu, Xingyu, et al.
Pubblicazione: (2026)

Security Threat Modeling for Emerging AI-Agent Protocols: A Comparative Analysis of MCP, A2A, Agora, and ANP
di: Anbiaee, Zeynab, et al.
Pubblicazione: (2026)

Secret Collusion among AI Agents: Multi-Agent Deception via Steganography
di: Motwani, Sumeet Ramesh, et al.
Pubblicazione: (2024)

BashArena: A Control Setting for Highly Privileged AI Agents
di: Kaufman, Adam, et al.
Pubblicazione: (2025)

Breaking the Protocol: Security Analysis of the Model Context Protocol Specification and Prompt Injection Vulnerabilities in Tool-Integrated LLM Agents
di: Maloyan, Narek, et al.
Pubblicazione: (2026)

Manipulating LLM Web Agents with Indirect Prompt Injection Attack via HTML Accessibility Tree
di: Johnson, Sam, et al.
Pubblicazione: (2025)

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

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

Bypassing Prompt Injection Detectors through Evasive Injections
di: Rahman, Md Jahedur, et al.
Pubblicazione: (2026)