Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 17 papers
Filter5

Attack type

Prompt injection17Poisoning and backdoors18Malicious use1

Attack technique

Encoding and obfuscation3Selection manipulation3

Defense type

Detection and filtering3Permissions and approval3Monitoring and auditing3

Impact

Unauthorized actions11Data and privacy leaks4

System type

AI Agent45coding agent16Browser and computer use1Multi-agent systems3

Attack surface / entry point

User input12Web pages and interfaces10Documents and messages7Tool responses8Tool and skill descriptions17Code repositories and dependencies15Memory writes4Images, audio and video5Distribution channels13

Layer

Application layer17

System component

MCP and skills14Monitors4Permissions and sandboxing4
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 240papers have no tags yet and are not included in filtered results.
Attack type：Prompt injectionAttack type：Malicious useSystem type：coding agentSystem type：Browser and computer useAttack surface / entry point：Tool and skill descriptionsClear all

Another 240papers do not have category tags yet and are not included in filtered results.

  3. Analysis and theoryarXiv 2610.06001

    AgentSpy: Observing AI agent behavior at the system call layer and detecting skill injection attacks

    提出 AgentSpy，在系统调用层观测 Agent 行为并检测技能注入

    arXiv
    2610.06001
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    gpt-5.6-terra, deepseek-v4-flash, glm-5.3-flash and others4
    Abstract提出了系统级智能体监控框架 AgentSpy，揭示出通过测试背后的非预期行为与注入风险。

    本研究提出了面向 AI 智能体的系统级外部监控与行为分析框架 AgentSpy。

    Paper analysisRead analysis
  4. DefensesarXiv 2610.05163

    Researchers construct segmented prompt injection on Claude Code and Codex and release a boundary action audit benchmark

    提出 PAA 方法，在边界动作前审计长流程智能体的分阶段提示注入

    arXiv
    2610.05163
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Code (v2.1.172, Claude Sonnet 5), Codex CLI (v0.144.0, GPT-5.5)
    Abstract论文针对长流程注入提出边界动作审计与 PAA，在保持低误阻的同时实现高拦截率。

    本研究针对长流程智能体在执行任务时面临的分阶段间接提示注入威胁，提出了在动作生效前进行拦截的边界动作审计框架与路径对齐归因方法。

    Paper analysisRead analysis
  5. DefensesarXiv 2610.03089

    COBRA: A Dual-LLM Architecture for Defending Computer-Using Agents Against Branch-Jumping Attacks

    提出 COBRA 双 LLM 架构，防御计算机使用 Agent 的分支转向攻击

    arXiv
    2610.03089
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Evaluated models
    gpt-5, UI-TARS-1.5-7B, claude-sonnet-5 and others6

    Abstract提出分支转向防御架构 COBRA，通过事前能力约束与边界代理消除分支转向攻击并兼顾效用。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.35912

    MMSkillRisk：多模态技能图像中的隐藏指令可诱导 Agent 执行未授权操作

    提出 MMSkillRisk 基准，评测多模态技能图像中的隐藏指令诱导 Agent 越权操作

    arXiv
    2609.35912
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.6-sol, Kimi-K2.6, Kimi-K3 and others6

    Abstract论文提出 MMSkillRisk 基准与 NCVA 攻击，揭示多模态技能中教学图像潜藏指令可引发未授权操作并与良性任务共存。

    Paper analysisRead analysis
  7. Evaluation and benchmarksarXiv 2609.32635

    TrustFork：显示身份如何劫持 LLM 智能体编排的权限

    提出 TRUSTFORK 基准，评测展示身份劫持多智能体编排权限

    arXiv
    2609.32635
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems, coding agent
    Evaluated models
    GPT-5.6-Sol, GPT-5.6-Luna, Kimi-K3 and others8

    Abstract论文提出了多智能体安全基准 TRUSTFORK，揭示展示身份如何劫持操作权威，并证明事后验证无法弥补已发生的执行破坏。

    Paper analysisRead analysis
  8. Risky behaviorarXiv 2609.30266

    研究显示 Claude Code、Codex 等本地智能体可轻易篡改自身执行轨迹

    测试编程智能体被诱导或自主篡改自身执行轨迹的行为

    arXiv
    2609.30266
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.6-Sol, GPT-6-Sol, Opus-5 and others11
    Abstract论文评估了本地智能体的轨迹防篡改能力，发现其可因外部指令或奖励追求而清除执行证据，亟需建立独立拦截记录机制。

    这是一项针对本地大模型智能体执行轨迹完整性与安全防篡改边界的实证研究。

    Paper analysisRead analysis
  9. AttacksarXiv 2609.18217

    研究：跨通道碎片化攻击可绕过 MCP 单通道防御

    提出跨通道分段攻击，利用 MCP 隐式信任外发敏感文件

    arXiv
    2609.18217
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-4o, GPT-4o-mini, GPT-5.4 and others15

    Abstract论文揭示了 MCP 多通道隐式信任层级，通过跨通道分段与协议级利用突破现有防御，展示了严重的凭证外发风险。

    Paper analysisRead analysis
  12. DefensesarXiv 2609.01677

    Skill-as-API：面向智能体软件工程的机密多智能体协作协议

    提出 Skill-as-API，隔离多智能体协作中的 skill 正文与系统提示词

    arXiv
    2609.01677
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems, coding agent
    Abstract协议把 skill 正文留在所有者侧，四层限制发现内容、权限期限和注入拼接。

    Skill-as-API 是面向智能体软件工程的协调协议，用来在跨组织调用时把 skill 的代码和系统提示词留在所有者进程。作者针对三条威胁：经通道或命名空间抽取专有方法，用注入让被调用模型输出系统提示词，以及协作结束后权限继续保留。

    Paper analysisRead analysis
  13. AttacksarXiv 2608.25776

    EvoMal attack uses self-poisoning of self-evolving coding agents to achieve worm-like propagation

    提出 EVOMAL，诱导自演化编码 Agent 在创建技能时自我投毒并蠕虫传播

    arXiv
    2608.25776
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Devstral-Small-2, Gemma-4-31B-IT, Qwen3-Coder-Next and others7

    Abstract论文指出了自演化智能体模仿检索代码的自我毒化漏洞，提出 EVOMAL 攻击与防御，阐明了供应链新风险。

    Paper analysisRead analysis
  14. 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
    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。

    Paper analysisRead analysis
  15. AttacksarXiv 2607.12340

    Research: LLM agent skill-name hallucination rates reach up to 62%, which can be exploited via squatting for supply-chain attacks

    提出技能名称抢注攻击，利用 Agent 虚构技能名实施供应链劫持

    arXiv
    2607.12340
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GLM-4.5-Air, GPT-5.4-mini, Claude Sonnet 4.6 and others12
    Abstract研究证实智能体虚构技能名具有高发生率与可预测性，现有防御难以兼顾安全与效用，亟需注册表结构性防护。

    本研究系统揭示了大模型智能体技能推荐场景下的虚构技能名称安全风险，将其形式化为一种新型供应链抢注攻击。

    Paper analysisRead analysis
  16. DefensesarXiv 2605.17380

    Uber proposes ADR enterprise-level agent detection system, deployed on 7,200 hosts for ten months

    提出企业级检测系统 ADR，识别编程 Agent 的提示注入与恶意 MCP

    arXiv
    2605.17380
    Published
    Layer
    Application layer
    Setting
    coding agent

    Abstract提出了端点因果遥测、两级在线分流与离线红队闭环的 ADR 系统，并在生产环境中验证了有效性。

    Paper analysisRead analysis
  17. AttacksarXiv 2602.14211

    SkillJect: Automatically generating skill-based prompt injection for skill-based agents

    提出 SkillJect，自动生成投毒 skill 诱导代码 Agent 执行隐藏脚本

    arXiv
    2602.14211
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude-4.6-Sonnet, GPT-5-mini, GLM-4.7 and others10
    AbstractSkillJect 用双通道和闭环精炼投毒 skill。

    SkillJect 是针对可安装 agent skill 的自动化提示注入框架。作者要解决的是：公开或手工放入 skill 的恶意指令常常被拒绝或忽略，而 skill 一旦安装就会被反复当作可信指导加载。

    Paper analysisRead analysis
You have reached the end