Skip to content

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

Oct 13 · Tuesday

All papers

Found 2 papers
Filter5

Attack surface / entry point

User input13Web pages and interfaces5Documents and messages7Tool responses14Tool and skill descriptions7Code repositories and dependencies1Images, audio and video0Distribution channels4Inference services and hardware1

Layer

Application layer2

System component

Memory4Guardrails and judges10Monitors2Reasoning chains3Permissions and sandboxing7Vision16Audio3Multimodal19
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Attack surface / entry point：Code repositories and dependenciesAttack surface / entry point：Images, audio and videoAttack surface / entry point：Inference services and hardwareLayer：Application layerSystem component：MonitorsClear all

Another 215papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2609.35659

    Tracekit：面向自主编码 Agent 的防篡改意图-推理-动作审计系统

    提出 Tracekit，审计编码 Agent 的意图、推理与动作并检测账本篡改

    arXiv
    2609.35659
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Code (claude -p, version 2.1), 独立评审（同一 CLI，无工具）
    AbstractTracekit 把意图、自述和动作写入可锚定的哈希链。

    Tracekit 是无第三方依赖的编码智能体审计层：把人类请求、模型自述和已执行动作并排写入可外部锚定的哈希链，执行前用正则门，事后用规则和独立模型互查。

    Paper analysisRead analysis
  2. DefensesarXiv 2609.01931

    Agent Flight Recorder：为长时程工具调用智能体提供链上锚定的防篡改审计追踪

    提出 Agent Flight Recorder，为长时程工具智能体构建链上锚定的防篡改审计追踪

    arXiv
    2609.01931
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Abstract语义事件加哈希链、Merkle 与链上锚定，用来在争议中核验长程智能体的动作序列。

    Agent Flight Recorder 把长程工具智能体的动作收成可被第三方独立核验的审计记录，面向运营方本身可能改写历史的争议。

    Paper analysisRead analysis
You have reached the end