DefensesarXiv 2610.02869
AgentTrap：面向自主渗透测试 Agent 的有状态反馈欺骗蜜罐
提出 AgentTrap 有状态蜜罐，欺骗并反击自主渗透测试 Agent
- arXiv
- 2610.02869
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- AIDA, Claude Code, PentestGPT and others8
- AttacksMalicious use
All papers
Filter5
Attack type
Attack technique
Impact
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.
Another 233papers do not have category tags yet and are not included in filtered results.
- AttacksarXiv 2609.33910
论文提出残留授权重放攻击：长期运行 Agent 的授权可跨任务复用
提出残余权限重放攻击，跨任务复用长期 Agent 的历史授权绕过确认
- arXiv
- 2609.33910
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- GPT-4.1, Gemini-3.1-Flash-Lite, Qwen3-14B and others6
- AttacksPrompt injection
- TechniqueDefense-targeted
- ComponentPermissions and sandboxing
Cave-Bench：虚假指控下 LLM 智能体最高 60.06% 的运行会破坏正确工作
提出 CAVE-BENCH，评测智能体在虚假指责下破坏已完成的正确工作
- arXiv
- 2609.32616
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- Claude-Sonnet-5, Claude-Opus-5, GPT-5.6-Sol and others14
- AttacksPrompt injection
- RiskDangerous agent actions
Abstract论文提出了评估智能体在虚假指责下破坏已有正确成果的 CAVE-BENCH，发现强模型易推翻已知证据，提出了门控缓解方案。
- AttacksarXiv 2609.01222
Researchers propose context privilege escalation attack, affecting 12 agent frameworks
提出针对 Agent Harness 的上下文特权提升攻击
- arXiv
- 2609.01222
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- GPT-5.5, GPT-5.4 mini, Claude-Sonnet-4.6 and others7
- AttacksPrompt injection
- TechniqueDormant triggers
- ComponentMemory
- AttacksarXiv 2608.27439
RedEvoAgent：以经验驱动技能演化的自动红队智能体
提出 RedEvoAgent，演化攻击技能对黑盒智能体做自动红队
- arXiv
- 2608.27439
- Published
- Layer
- Application layer
- Setting
- coding agent
- Attacker
- Black-box
- Evaluated models
- MiniMax-M2.5, DeepSeek-V4-Flash, Qwen3.5-35B
- AttacksJailbreaks
- TechniqueAutomated search
AbstractRedEvoAgent 用验证棘轮演化可读攻击技能，在两种产品级 harness 上多数设置高于固定工具与红队基线。
RedEvoAgent 是面向产品级 black-box 智能体的自动红队方法，用一份可演化、人类可读的攻击技能编排多个越狱工具。。
- DefensesarXiv 2608.21101
ClawSentry：面向自主 LLM Agent 的渐进式多层安全监控网关
提出 ClawSentry 多层监控网关，在技能准入和运行时拦截恶意技能与隐藏指令
- arXiv
- 2608.21101
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- Codex/GPT-5.4, Codex/GPT-5.5, Claude Code/GLM-5.1 and others6
- DefensesMonitoring and auditing
- AttacksPoisoning and backdoors
- ComponentMCP and skills
AbstractClawSentry 用包准入、三级运行时门控和反绕过，压低 skill 攻击并保住干净任务。
ClawSentry 是挂在智能体运行时之外的安全监督网关，用来同时盖住 skill 执行的多个生命周期边界，并把已被拒绝的效果记到会话里。
The researchers propose SkillMisevo-Bench to quantify the risk of skill mis-evolution in self-improving agents.
提出 SkillMisevo-Bench，量化自改进 Agent 的技能误演化
- arXiv
- 2608.12851
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- Claude Code, Codex, Hermes and others4
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMemory
- DefensesarXiv 2607.21642
CARE: Pre-execution command validation for LLM agents that execute shell commands
Proposes a static-first validator, CARE, that decides before dispatch whether an agent's shell command may be executed
- arXiv
- 2607.21642
- Published
- Layer
- Application layer
- Setting
- coding agent
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
- AttacksarXiv 2607.11698
AHA 用自动研究循环发现生产 Agent 漏洞概念
提出 AHA 自动科研循环，发现生产 Agent 的可复用漏洞概念
- arXiv
- 2607.11698
- Published
- Layer
- Application layer
- Setting
- coding agent
- Attacker
- Black-box
- Evaluated models
- Minimax-M2.7, Kimi-K2.6, Deepseek-V4-Pro and others5
- AttacksPrompt injection
- TechniqueAutomated search
Abstract论文提出基于可证伪科研循环的 AHA 框架，从生产级智能体中挖掘并沉淀漏洞概念图，提升攻击复用率并指导防御加固。
AHA 是一项针对生产级 LLM 智能体自动化红队测试的研究，旨在解决传统方法仅保留具体攻击载荷而缺乏成因解释、导致跨版本复用困难且无法指导修复的问题。
- DefensesarXiv 2606.22673
Agent MechSuits：面向多轮 CLI Agent 的机制子空间安全引导
提出 Agent MechSuits，对多轮 CLI Agent 做步级检测与子空间转向
- arXiv
- 2606.22673
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- LLaMA-3.1-8B, Qwen-2.5-7B, Gemma-2-9B
- DefensesInference-time intervention
- AttacksMalicious use
- ComponentMonitors
Abstract白盒步级探针加稀疏转向可压低三模型恶意动作，同时提高生成崩溃率。
Agent MechSuits 是面向多轮 CLI 智能体的白盒、步级表示防御，并配有可执行轨迹基准 MAS。
- AttacksarXiv 2605.30883
TRACE：面向 LLM Agent 的任务感知自适应自进化越狱框架
提出 TRACE 框架，用分解与可演化多轮策略诱导智能体执行有害工作流
- arXiv
- 2605.30883
- Published
- Layer
- Application layer
- Setting
- coding agent
- Evaluated models
- Codex, Claude Code, GPT-5.5 and others6
- AttacksJailbreaks
- TechniqueMulti-turn escalation
AgentHazard：评估计算机使用智能体有害行为的基准
提出 AgentHazard 基准，评测计算机使用智能体的执行层有害行为
- arXiv
- 2604.02947
- Published
- Layer
- Application layer
- Setting
- Browser and computer use, coding agent
- Evaluated models
- Qwen2.5-72B-Instruct, Qwen2.5-Coder-32B-Instruct, Qwen3-32B and others12
- AttacksMalicious use
AbstractAgentHazard 评估了计算机使用智能体的执行安全，作者认为模型层对齐无法可靠保障智能体层面的安全性。
AgentHazard 是针对计算机使用智能体执行层安全构建的评估基准，重点探究多步工具调用下局部合理动作累积引发的危害。
You have reached the end