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

Oct 12 · Monday

All papers

Found 29 papers

Filter5

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 232papers have no tags yet and are not included in filtered results.

System type：Models and APIs System type：coding agent System type：science agent System type：Multi-agent systems System component：MCP and skills Clear all

Another 232papers do not have category tags yet and are not included in filtered results.

Evaluation and benchmarksarXiv 2610.11647Oct 8
One More Skill Installed: How Co-Installed Skills in Coding Agents Conflict
评测编码 Agent 中相似技能共同安装对已装技能核心功能的替代
arXiv
2610.11647
Published
Oct 8
Layer
Application layer
Setting
coding agent
Evaluated models
claude-sonnet-4-6, claude-haiku-4-5-20251001, claude-opus-5 and others6
- RiskDangerous agent actions
- ComponentMCP and skills
Abstract冲突藏在任务成功后面。
作者对共同安装、做同一件事的良性 skill 做实证研究，用核心功能是否还在，而不是任务过不过，来看冲突。。
Paper analysisRead analysis
Analysis and theoryarXiv 2610.11169Oct 8
研究重建 GitHub 上 Agent 技能的复制供应链网络
重建 Agent 技能复制网络并建模源选择以指导安全审计
arXiv
2610.11169
Published
Oct 8
Layer
Application layer
Setting
coding agent
- ComponentMCP and skills
Abstract论文重构了 GitHub 智能体技能复制网络，发现复制集中于少数仓库且很少随源更新，并提出了代码库审计推荐模型。
Paper analysisRead analysis
AttacksarXiv 2610.10612Oct 7
PyCache Trap 利用 Python 缓存替换绕过七款 Agent Skill 扫描器
提出 PyCache Trap，用字节码缓存替换绕过 Agent 技能扫描器
arXiv
2610.10612
Published
Oct 7
Layer
Application layer
Setting
coding agent
Evaluated models
CISCO, HSS, AUDITOR and others8
- DefensesDetection and filtering
- AttacksDetection evasion
- TechniqueDefense-targeted
- ComponentMCP and skills
Abstract论文剖析智能体技能审查与执行脱节盲区，提出 PyCache Trap 攻击并构建基于可信复现的 EAV 防御。
这篇论文研究了智能体技能（Agent Skills）安全扫描中的“审查–执行脱节”问题，并提出了对应的攻击验证与防御方案。
Paper analysisRead analysis
AttacksarXiv 2610.08098Oct 6
CORSA：面向技能路由器的技能注入攻击可跨路由器与 LLM 迁移
提出 CORSA，按任务簇优化技能注入的检索与执行
arXiv
2610.08098
Published
Oct 6
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5.4, DeepSeek-V4-Pro, Qwen3.8-27B and others5
- AttacksPoisoning and backdoors
- TechniqueSelection manipulation
- ComponentMCP and skills
Paper analysisRead analysis
Analysis and theoryarXiv 2610.06001Oct 5
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
Oct 5
Layer
Application layer
Setting
coding agent
Evaluated models
gpt-5.6-terra, deepseek-v4-flash, glm-5.3-flash and others4
- DefensesDetection and filtering
- AttacksPoisoning and backdoors
- ComponentMCP and skills
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
AttacksarXiv 2610.00430Oct 1
Memetic Trojans：利用社交传染在 LLM Agent 网络中传播对抗载荷
提出 memetic trojans，借助社交传染在 Agent 网络中传播对抗载荷
arXiv
2610.00430
Published
Oct 1
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
gpt-oss-120B, deepseek-v4.1-flash, qwen-32B and others5
- AttacksPoisoning and backdoors
- ComponentMCP and skills
Abstractmemetic trojans 用内生社会传染携带载荷。
Memetic trojans 是一类网络介导攻击：对抗载荷搭在智能体本来就会转发和点赞的内容上扩散，转发者不必被攻陷，也不必执行“请继续复制”的指令。
Paper analysisRead analysis
AttacksarXiv 2610.00392Sep 30
A2A-TIBA：针对 A2A 协议 LLM 智能体的间接提示注入攻击与三层同构攻防模型
提出 A2A-TIBA，经 A2A 对等任务间接注入并植入常驻回调
arXiv
2610.00392
Published
Sep 30
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
MiniMax-M3, deepseek-v4-pro, glm-5.3 and others8
- AttacksPrompt injection
- TechniqueMulti-turn escalation
- ComponentMCP and skills
AbstractA2A-TIBA 一次植入后绕过智能体，GDA 用四类结果定位防线，ELA-ITL 把信封标注列为第三层。
A2A-TIBA 把间接提示注入和绕过放进 A2A：gray-box 攻击者已能投递任务，诱导目标部署常驻回调程序后，下令和外传不再经过智能体。
Paper analysisRead analysis
AttacksarXiv 2609.39352Sep 30
研究者提出解耦铺垫与执行的 LLM Agent 技能投毒攻击
提出 CoordPoison，解耦技能投毒的借口与动作
arXiv
2609.39352
Published
Sep 30
Layer
Application layer
Setting
coding agent
Evaluated models
DeepSeek-V4-Flash, GLM-5.3-Flash, Claude-Sonnet-4.6 and others5
- AttacksPoisoning and backdoors
- TechniqueAutomated search
- ComponentMCP and skills
AbstractCoordPoison 外置借口。
CoordPoison 是一种 Skill 供应链投毒方法，把“为何执行”和“执行什么”拆到两个 Skill。。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.35912Sep 28
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
Sep 28
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5.6-sol, Kimi-K2.6, Kimi-K3 and others6
- AttacksPrompt injection
- TechniqueCross-modal payloads
- ComponentMCP and skills
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
AttacksarXiv 2609.32400Sep 26
SkillDRE：用预执行与运行时双阶段反馈自动演化恶意 Agent 技能
提出 SkillDRE，用预执行与运行时双阶段反馈演化恶意 Agent 技能
arXiv
2609.32400
Published
Sep 26
Layer
Application layer
Setting
coding agent
Evaluated models
DeepSeek-V4-Pro, GLM-5.2, Qwen3.5-397B-A17B and others4
- AttacksPoisoning and backdoors
- TechniqueAutomated search
- ComponentMCP and skills
Paper analysisRead analysis
Risky behaviorarXiv 2609.30266Sep 25
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
Sep 25
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5.6-Sol, GPT-6-Sol, Opus-5 and others11
- AttacksMalicious use
- RiskDeception and scheming
- ComponentMonitors
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
Evaluation and benchmarksarXiv 2609.28915Sep 24
研究提出 ACE 语料库，评估内核级证据对 Agent 安全检测的有效性
构建 ACE 语料，评测内核级证据对 Agent 攻击检测的有效性
arXiv
2609.28915
Published
Sep 24
Layer
Application layer
Setting
coding agent
Evaluated models
AdaBoost, XGBoost, TabPFN and others10
- AttacksPrompt injection
- ComponentMCP and skills
AbstractACE 用配对的内核与应用层证据表明，syscall 痕迹可判别智能体攻击，两层组合通常更好。
ACE 是一份把智能体会话的内核 syscall 痕迹与应用层清单、转录配对起来的检测语料和基准。。
Paper analysisRead analysis
AttacksarXiv 2609.18217Sep 16
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
Sep 16
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-4o, GPT-4o-mini, GPT-5.4 and others15
- AttacksPrompt injection
- TechniqueEncoding and obfuscation
- ComponentMCP and skills
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
DefensesarXiv 2609.14079Sep 13
SkillSecurer：检测并修补 AI Agent 技能中的提示注入漏洞
提出 SkillSecurer，检测并修补 Agent 技能中的提示注入漏洞
arXiv
2609.14079
Published
Sep 13
Layer
Application layer
Setting
coding agent
Evaluated models
Claude Sonnet 5, Gemini 3.6 Flash, GPT-5.6 Sol and others10
- DefensesDetection and filtering
- AttacksPrompt injection
- ComponentMCP and skills
Paper analysisRead analysis
DefensesarXiv 2609.01487Sep 2
Defense-as-Skill：为技能增强型 Agent 演化运行时护栏技能
提出 SkillSonar，把运行时防护做成可演化技能以拦截恶意技能
arXiv
2609.01487
Published
Sep 2
Layer
Application layer
Setting
coding agent
Evaluated models
GLM-5, GPT-5.4, Claude Haiku 4.5 and others4
- DefensesInstruction–data separation
- AttacksPoisoning and backdoors
- ComponentMCP and skills
Paper analysisRead analysis
AttacksarXiv 2609.01222Sep 1
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
Sep 1
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
Paper analysisRead analysis
DefensesarXiv 2609.01677Sep 1
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
Sep 1
Layer
Application layer
Setting
Multi-agent systems, coding agent
- DefensesInstruction–data separation
- AttacksExtraction and theft
- ComponentMCP and skills
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
DefensesarXiv 2608.30083Aug 31
研究者提出零知识谓词证明网关，在 MCP 与 Agent2Agent 上实现 Agent 间数据最小化
提出 Proof Gateway，用零知识谓词证明实现 Agent 间数据最小化
arXiv
2608.30083
Published
Aug 31
Layer
Application layer
Setting
Multi-agent systems
- AttacksPrompt injection
- ComponentMCP and skills
Abstract作者将 source integrity 而不是证明开销视为绑定约束。
Proof Gateway 是一套已运行的智能体间数据最小化机制：用治理谓词的零知识证明代替原始值和自然语言自述。
Paper analysisRead analysis
AttacksarXiv 2608.25776Aug 26
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
Aug 26
Layer
Application layer
Setting
coding agent
Evaluated models
Devstral-Small-2, Gemma-4-31B-IT, Qwen3-Coder-Next and others7
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMCP and skills
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
AttacksarXiv 2608.21929Aug 22
SkillBloat：通过技能注入对编码 Agent 发起 token 放大攻击
提出 SKILLBLOAT，改写技能文档放大编码 Agent token 消耗
arXiv
2608.21929
Published
Aug 22
Layer
Application layer
Setting
coding agent
Evaluated models
GLM-4.7-Flash, GLM-5, GPT-5.4 Mini and others4
- AttacksDenial of service
- TechniqueAutomated search
- ComponentMCP and skills
Paper analysisRead analysis

AI Safety Daily

Read the latest issueOne issue per day, updated as events develop

All papers

Attack type

Attack technique

Defense type

Impact

System type

Attack surface / entry point

Layer

System component

One More Skill Installed: How Co-Installed Skills in Coding Agents Conflict

研究重建 GitHub 上 Agent 技能的复制供应链网络

PyCache Trap 利用 Python 缓存替换绕过七款 Agent Skill 扫描器

CORSA：面向技能路由器的技能注入攻击可跨路由器与 LLM 迁移

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

Memetic Trojans：利用社交传染在 LLM Agent 网络中传播对抗载荷

A2A-TIBA：针对 A2A 协议 LLM 智能体的间接提示注入攻击与三层同构攻防模型

研究者提出解耦铺垫与执行的 LLM Agent 技能投毒攻击

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

SkillDRE：用预执行与运行时双阶段反馈自动演化恶意 Agent 技能

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

研究提出 ACE 语料库，评估内核级证据对 Agent 安全检测的有效性

研究：跨通道碎片化攻击可绕过 MCP 单通道防御

SkillSecurer：检测并修补 AI Agent 技能中的提示注入漏洞

Defense-as-Skill：为技能增强型 Agent 演化运行时护栏技能

Researchers propose context privilege escalation attack, affecting 12 agent frameworks

Skill-as-API：面向智能体软件工程的机密多智能体协作协议

研究者提出零知识谓词证明网关，在 MCP 与 Agent2Agent 上实现 Agent 间数据最小化

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

SkillBloat：通过技能注入对编码 Agent 发起 token 放大攻击