Skip to content

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

Oct 12 · Monday

All papers

Found 2 papers
Filter3

Impact

Harmful content2Unauthorized actions41Data and privacy leaks14Output manipulation8Availability and reliability8Oversight failure16Increased misuse capability5

System type

LLM applications7AI Agent23coding agent2science agent4Multi-agent systems7

Layer

Application layer2
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 261papers have no tags yet and are not included in filtered results.
Impact：Harmful contentSystem type：coding agentLayer：Application layerClear all

Another 261papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2607.03968

    研究揭示 IDE 编程 Agent 的工作流级越狱：GitHub Copilot 四后端 816/816 生成有害内容

    提出工作流级越狱，让编程 Agent 在开发任务中写入有害示例

    arXiv
    2607.03968
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Sonnet 4.6, Claude Haiku 4.5, Gemini 3.1 Pro and others4
    Abstract作者报告 Copilot 后端会在编码工作流中生成其在单轮条件下通常拒绝的有害答案。

    这是一项生产 IDE 编码智能体的工作流越狱研究，受测对象是 Copilot 的四个模型后端，而非评测流水线中的名义目标。

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2510.01359

    JAWS-Bench：研究者用三种工作区场景系统越狱代码 Agent

    提出 JAWS-Bench，评测代码 Agent 在三级工作区中的越狱与恶意代码可执行性

    arXiv
    2510.01359
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-4.1, GPT-o1, DeepSeek-R1 and others8

    Abstract论文提出了评估代码智能体端到端恶意代码越狱的 JAWS-Bench，发现多文件与智能体推理会明显放大可部署危害。

    Paper analysisRead analysis
You have reached the end