Copertina

Servant, Stalker, Predator: How An Honest, Helpful, And Harmless (3H) Agent Unlocks Adversarial Skills

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autore principale:	Noever, David
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Cryptography and Security Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Language Models And A Second Opinion Use Case: The Pocket Professional
di: Noever, David
Pubblicazione: (2024)

Hallucinating AI Hijacking Attack: Large Language Models and Malicious Code Recommenders
di: Noever, David, et al.
Pubblicazione: (2024)

Towards Safe and Honest AI Agents with Neural Self-Other Overlap
di: Carauleanu, Marc, et al.
Pubblicazione: (2024)

HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?
di: Jiang, Yukun, et al.
Pubblicazione: (2026)

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

Understanding LLM Behavior When Encountering User-Supplied Harmful Content in Harmless Tasks
di: Chu, Junjie, et al.
Pubblicazione: (2026)

HonestCyberEval: An AI Cyber Risk Benchmark for Automated Software Exploitation
di: Ristea, Dan, et al.
Pubblicazione: (2024)

SkillTester: Benchmarking Utility and Security of Agent Skills
di: Wang, Leye, et al.
Pubblicazione: (2026)

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

SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agents
di: Jia, Xiaojun, et al.
Pubblicazione: (2026)

SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems
di: Feng, Yunhao, et al.
Pubblicazione: (2026)

Context Matters: Repository-Aware Security Analysis of the Agent Skill Ecosystem
di: Holzbauer, Florian, et al.
Pubblicazione: (2026)

SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills
di: Hou, Yinghan, et al.
Pubblicazione: (2026)

AgentTrap: Measuring Runtime Trust Failures in Third-Party Agent Skills
di: Zhuang, Haomin, et al.
Pubblicazione: (2026)

Detecting Adversarial Fine-tuning with Auditing Agents
di: Egler, Sarah, et al.
Pubblicazione: (2025)

Technical Report: Exploring the Emerging Threats of the Agent Skill Ecosystem
di: Beurer-Kellner, Luca, et al.
Pubblicazione: (2026)

Structured Security Auditing and Robustness Enhancement for Untrusted Agent Skills
di: Lv, Lijia, et al.
Pubblicazione: (2026)

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

Living Off the LLM: How LLMs Will Change Adversary Tactics
di: Oesch, Sean, et al.
Pubblicazione: (2025)

Towards Secure Agent Skills: Architecture, Threat Taxonomy, and Security Analysis
di: Li, Zhiyuan, et al.
Pubblicazione: (2026)

Proteus: A Self-Evolving Red Team for Agent Skill Ecosystems
di: Zhou, Zhaojiacheng
Pubblicazione: (2026)

RouteGuard: Internal-Signal Detection of Skill Poisoning in LLM Agents
di: Xiao, Wenjie, et al.
Pubblicazione: (2026)

Latent Fusion Jailbreak: Blending Harmful and Harmless Representations to Elicit Unsafe LLM Outputs
di: Xing, Wenpeng, et al.
Pubblicazione: (2025)

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

SkCC: Portable and Secure Skill Compilation for Cross-Framework LLM Agents
di: Ouyang, Yipeng, et al.
Pubblicazione: (2026)

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

WARD: Adversarially Robust Defense of Web Agents Against Prompt Injections
di: Cao, Tri, et al.
Pubblicazione: (2026)

How stealthy is stealthy? Studying the Efficacy of Black-Box Adversarial Attacks in the Real World
di: Panebianco, Francesco, et al.
Pubblicazione: (2025)

ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers
di: Liu, Songyang, 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)

DECEPTICON: How Dark Patterns Manipulate Web Agents
di: Cuvin, Phil, et al.
Pubblicazione: (2025)

Explanation as a Watermark: Towards Harmless and Multi-bit Model Ownership Verification via Watermarking Feature Attribution
di: Shao, Shuo, et al.
Pubblicazione: (2024)

Doppelganger Method: Breaking Role Consistency in LLM Agent via Prompt-based Transferable Adversarial Attack
di: Kang, Daewon, et al.
Pubblicazione: (2025)

When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems
di: Wang, Su, et al.
Pubblicazione: (2026)

The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
di: Ding, Xuwei, et al.
Pubblicazione: (2026)

Next-Generation Phishing: How LLM Agents Empower Cyber Attackers
di: Afane, Khalifa, et al.
Pubblicazione: (2024)

Behavioral Integrity Verification for AI Agent Skills
di: Wu, Yuhao, et al.
Pubblicazione: (2026)

AccLock: Unlocking Identity with Heartbeat Using In-Ear Accelerometers
di: Wang, Lei, et al.
Pubblicazione: (2026)

From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection
di: Wang, Haowei, et al.
Pubblicazione: (2024)

E-PhishGen: Unlocking Novel Research in Phishing Email Detection
di: Pajola, Luca, et al.
Pubblicazione: (2025)