跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.10843 · v1

Agents of Chaos: A Systematization of Knowledge (SoK) on Malicious Attacks on Agentic AI Systems

首次提交 2026-11-08 00:42
版本更新 2026-11-08 00:42
雷达观测 2026-11-10 11:14
arXiv ↗PDF ↗

摘要

Agentic AI systems extend large language models with tools, memory, retrieval, browser or computer-use interfaces, code execution, external services, and multi-agent coordination. Although these capabilities enable agents to act beyond text generation and perform tasks, they also create attack paths through which adversaries can exploit untrusted content, deceptive metadata, poisoned state, compromised skills, mutable dependencies, or malicious code. Such attacks can manipulate planning and tool use, corrupt memory, abuse code-execution environments, and trigger harmful downstream effects. In this paper, we present a systematic review of malicious attacks targeting agentic AI systems. We screened 186 candidate sources and retained 98. Following independent coding and resolver adjudication, we identified 284 atomic attacks, 452 atomic safeguards, and 660 explicit attack--safeguard relationships. We organize attacks into eight causal-mechanism classes while retaining insertion points, activation conditions, abused capabilities, objectives, impacts, persistence, and evidence as orthogonal dimensions. Our analysis shows that tool, MCP, and action-layer attacks account for the largest share of atomic attacks, while the prompt-injection and action-hijacking class has the broadest paper coverage. We identify distinct class-specific activation patterns and separation between insertion and later activation. Safeguard coverage remains uneven: only 278 of 660 relationships are directly assessed, and strong coverage is uncommon. These findings support lifecycle-aware threat modeling, agent evaluation, tool governance, and defense-in-depth security for agentic AI systems.

作者备注:Accepted at the Workshop on Agentic AI in Offensive and Defensive Cyber Operations (AIDC 2026), co-located with ACSAC 2026

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-08 00:42 · 最新观测版本

Agents of Chaos: A Systematization of Knowledge (SoK) on Malicious Attacks on Agentic AI Systems

Agentic AI systems extend large language models with tools, memory, retrieval, browser or computer-use interfaces, code execution, external services, and multi-agent coordination. Although these capabilities enable agents to act beyond text generation and perform tasks, they also create attack paths through which adversaries can exploit untrusted content, deceptive metadata, poisoned state, compromised skills, mutable dependencies, or malicious code. Such attacks can manipulate planning and tool use, corrupt memory, abuse code-execution environments, and trigger harmful downstream effects. In this paper, we present a systematic review of malicious attacks targeting agentic AI systems. We screened 186 candidate sources and retained 98. Following independent coding and resolver adjudication, we identified 284 atomic attacks, 452 atomic safeguards, and 660 explicit attack--safeguard relationships. We organize attacks into eight causal-mechanism classes while retaining insertion points, activation conditions, abused capabilities, objectives, impacts, persistence, and evidence as orthogonal dimensions. Our analysis shows that tool, MCP, and action-layer attacks account for the largest share of atomic attacks, while the prompt-injection and action-hijacking class has the broadest paper coverage. We identify distinct class-specific activation patterns and separation between insertion and later activation. Safeguard coverage remains uneven: only 278 of 660 relationships are directly assessed, and strong coverage is uncommon. These findings support lifecycle-aware threat modeling, agent evaluation, tool governance, and defense-in-depth security for agentic AI systems.

查看此版本解读 →查看此版本原文 ↗