Skip to content

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

Oct 12 · Monday

All papers

Found 11 papers
Filter3

Attack type

Jailbreaks5Prompt injection11Poisoning and backdoors4

Defense type

Inference-time intervention3Detection and filtering14Instruction–data separation5Permissions and approval11Monitoring and auditing8

Impact

Unauthorized actions9

System type

AI Agent11

Attack surface / entry point

User input11Web pages and interfaces11Documents and messages9Tool responses21Tool and skill descriptions10Retrieval corpus3Memory writes7Inter-agent messages4Distribution channels4

Layer

Application layer11

System component

MCP and skills4Memory3Guardrails and judges3Permissions and sandboxing7
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 265papers have no tags yet and are not included in filtered results.
Attack type：Prompt injectionDefense type：Permissions and approvalAttack surface / entry point：User inputClear all

Another 265papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2610.07657

    DEFER1 多智能体防御研究报告受控测试中的攻击成功率下降

    实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击

    arXiv
    2610.07657
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen3-235B-A22B-Instruct-2507, gpt-oss-120b, Llama-4-Scout-17B-16E-Instruct and others5
    AbstractDEFER 把确定性检查放在 judge 之前。

    DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。

    Paper analysisRead analysis
  3. DefensesarXiv 2610.02432

    论文提出 LLM 输入扰动鲁棒性度量，并发布 ASA 攻击与 AttestMCP 防护

    提出 ASA 与 AttestMCP，度量并降低 LLM 对输入序列扰动的敏感

    arXiv
    2610.02432
    Published
    Layer
    Application layer
    Setting
    LLM applications, AI Agent
    Attacker
    Black-box
    Evaluated models
    Gemma-3-27B-Instruct, Gemma-3-4B-Instruct, Llama-3.2-3B-Instruct and others12
    Paper analysisRead analysis
  4. DefensesarXiv 2609.24165

    APEXA 为同步辐射数据处理的 LLM 多智能体加装执行完整性护栏

    提出 APEXA，在工具层拒绝无执行支撑的结果并拦截电机命令

    arXiv
    2609.24165
    Published
    Layer
    Application layer
    Setting
    science agent
    Evaluated models
    GPT-5-mini, GPT-5.4, Claude Opus 4.7 and others4
    Abstract作者称执行是否发生要由工具层核对，而不能只靠提示词。

    APEXA 是先进光子源上的数据还原框架，用工具层代码约束语言模型智能体的工具调用。

    Paper analysisRead analysis
  5. DefensesarXiv 2609.22573

    研究提出面向企业 MCP 的零信任授权与工具发现方案

    提出企业 MCP 零信任授权与发现扩展，在架构上拒绝提示注入下的越权调用

    arXiv
    2609.22573
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    gpt-5, claude-sonnet-4-6, claude-opus-4-7 and others6
    Abstract同一份按工具声明同时管 MCP 的发现、列表和调用，使注入不能扩大已有凭证的权限。

    作者把零信任放在 MCP 工具调用边界：智能体即使被提示注入，也不能调用其凭证范围之外的工具。问题来自三点部署性质。调用意图会随不可信上下文改变。

    Paper analysisRead analysis
  8. DefensesarXiv 2609.29647

    AgentKernel：面向 Agent 的可信原生操作系统内核

    提出 AgentKernel，用四支柱强制中介 Agent 的输入、工具与记忆

    arXiv
    2609.29647
    Published
    Layer
    Application layer
    Setting
    AI Agent
    AbstractAgentKernel 是四支柱智能体 OS 设计，安全论证未附带实测数字。

    AgentKernel 是把安全做成组织原则的智能体 OS 架构，不是带跑分的攻防评测。

    Paper analysisRead analysis
  9. 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
You have reached the end