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

Oct 12 · Monday

All papers

Found 34 papers

Filter4

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.

Attack type：Prompt injection Impact：Data and privacy leaks Impact：Availability and reliability Layer：Application layer Clear all

Another 233papers do not have category tags yet and are not included in filtered results.

AttacksarXiv 2610.12524Oct 8
An exploration of attacks on AI agent systems
重测并扩展对 AI Agent 的提示注入，并用 Agent Card 欺骗操纵 A2A 选择
arXiv
2610.12524
Published
Oct 8
Layer
Application layer
Setting
AI Agent
Evaluated models
Ollama Mistral 0.9.5, llama3-8b-8192, gemini-2.0-flash-lite and others5
- AttacksPrompt injection
- TechniqueEncoding and obfuscation
- ComponentMCP and skills
Abstract单智能体过滤与分类可被格式或措辞绕过，A2A 选卡还会偏向名称和描述。
作者在自建的 LangChain、AutoGen 和 A2A 系统上重测提示注入，并改 Agent Card 看选择是否偏移。摘要称试验主要凸显单智能体弱点，同时把攻击扩到多智能体。
Paper analysisRead analysis
DefensesarXiv 2610.08773Oct 7
AdvSim2Real 用任务与注入对手共同演化训练网页智能体
提出 AdvSim2Real，协同训练网页智能体抵御页面注入
arXiv
2610.08773
Published
Oct 7
Layer
Application layer
Setting
web agent
Evaluated models
Qwen3.5-4B
- DefensesModel hardening
- AttacksPrompt injection
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.08871Oct 6
CredLeak-Bench
提出 CredLeak-Bench，评测 Agent 面对钓鱼的凭据泄露与恢复
arXiv
2610.08871
Published
Oct 6
Layer
Application layer
Setting
web agent
Evaluated models
Claude Sonnet 5, Claude Opus 4.8, Gemini 3.6 Flash and others7
- AttacksPrompt injection
Paper analysisRead analysis
AttacksarXiv 2610.04195Oct 3
MemLeak research: multi-tenant agents sharing vector memory stores have cross-user memory leakage
提出 MemLeak，评估多租户 Agent 共享向量记忆的跨用户泄露
arXiv
2610.04195
Published
Oct 3
Layer
Application layer
Setting
AI Agent
Evaluated models
Gemini 2.5 Flash, Claude Sonnet 4.5, all-MiniLM-L6-v2 and others5
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMemory
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.03448Oct 2
Study retests 15 prompt injection detectors: benchmark scores poorly predict agent deployment performance
重测提示注入检测器在 Agent 工具输出上的检出与任务阻断
arXiv
2610.03448
Published
Oct 2
Layer
Application layer
Setting
AI Agent
Evaluated models
Horizon-Labs, Wolf Defender, Prismor-1.5B and others15
- DefensesDetection and filtering
- AttacksPrompt injection
- ComponentGuardrails and judges
Abstract论文揭示公开基准无法反映智能体检测器表现，检测能力源于输入格式相似而非通用防御，建议基于工具输出与低误报评测。
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
DefensesarXiv 2610.00371Sep 30
FlowReview：以授权配对评测控制多智能体组合信息流
提出 FlowReview，用授权配对与确定性提交门控控制多智能体组合信息流
arXiv
2610.00371
Published
Sep 30
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
Sonnet 4.5, Haiku 4.5, Qwen3-32B and others5
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract配对标准要求拒绝禁止使用并完成授权使用，FlowReview 把三项审查接到执行。
这篇工作把多智能体安全写成一件必须同时成立的事：拦住组合信息所启用的禁止使用，并完成同一场景里任务所需的授权使用。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.34790Sep 28
CoSec：面向社区场景的 LLM Agent 安全基准
用 CoSec 评测社区 Agent 是否越界披露受保护信息
arXiv
2609.34790
Published
Sep 28
Layer
Application layer
Setting
AI Agent
Evaluated models
OpenClaw with Gemini 3.1 Pro, OpenClaw with Qwen 3.8 Max, OpenClaw with MiniMax M2.5 and others13
- AttacksPrompt injection
- ComponentMemory
Paper analysisRead analysis
DefensesarXiv 2609.33371Sep 27
论文提出保险库中介执行边界，评估 LLM Agent 的 API 凭据处理风险
提出并评测保险库中介执行边界，隔离 Agent 的 API 凭据
arXiv
2609.33371
Published
Sep 27
Layer
Application layer
Setting
AI Agent
Evaluated models
Corvic Security Vault
- DefensesInstruction–data separation
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract作者称，凭据分离可减少披露路径，但权限、操作授权和配置验证仍需独立控制。
这是一项智能体 API 凭据威胁分析及 Vault 执行边界的功能评测。作者关心的不是模型能否承诺保密，而是认证材料是否会进入模型、日志、记忆和执行代码。
Paper analysisRead analysis
DefensesarXiv 2609.30824Sep 25
基于加密绑定身份验证能力令牌协调分布式 AI 智能体的提案
提出身份绑定且委托只衰减的能力令牌，用上下文外钱包约束智能体
arXiv
2609.30824
Published
Sep 25
Layer
Application layer
Setting
Multi-agent systems
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract用只衰减的身份绑定令牌和上下文外钱包，约束代表人类行动的分布式智能体。
代表人类行动的分布式智能体，需要一种不把密钥放进语言模型上下文的授权方式。
Paper analysisRead analysis
AttacksarXiv 2609.27542Sep 23
Researchers use control token injection to delete gpt-oss-20b's reasoning chain and bypass chain-of-thought monitoring
提出控制标记注入，清空工具 Agent 的思维链以绕过推理监控
arXiv
2609.27542
Published
Sep 23
Layer
Application layer
Setting
AI Agent
Evaluated models
openai/gpt-oss-20b, google/gemma-4-26B-A4B-it, deepseek-ai/DeepSeek-R1-Distill-Qwen-7B and others4
- AttacksPrompt injection
- TechniqueConversation structure manipulation
- ComponentReasoning chains
Abstract论文阐明控制标记注入可抑制 CoT 使监控失效并促成拒绝绕过，且工具执行受解析器宽容度影响。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.19140Sep 17
AgentLSD：在对抗性任务污染下评测 AI 安全 Agent
评测安全 Agent 在对抗性任务污染下的解题成功与开销
arXiv
2609.19140
Published
Sep 17
Layer
Application layer
Setting
web agent
Evaluated models
Gemma-4 31B, DeepSeek-V4 Flash, GPT-OSS 120B and others6
- AttacksPrompt injection
AbstractAgentLSD 用配对 web CTF 表明，欺骗性环境证据会改变安全智能体的解题率与工作量。
AgentLSD 是一个对照框架，用来测量安全智能体在任务本身不变时，对对抗性任务污染的反应。
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
Evaluation and benchmarksarXiv 2609.15017Sep 14
PIDS-Bench：在过度防御、混淆与分布偏移下评测提示注入检测器
提出 PIDS-Bench，评测提示注入检测器在过度防御与分布偏移下的表现
arXiv
2609.15017
Published
Sep 14
Layer
Application layer
Setting
LLM applications
Evaluated models
TF-IDF + logistic regression, DistilBERT, DeBERTa-v3-FT and others7
- DefensesDetection and filtering
- AttacksPrompt injection
- ComponentGuardrails and judges
AbstractPIDS-Bench 显示高分布内 F1 仍会过度拦截外部来源的安全相关良性输入。
PIDS-Bench 是一个冻结的检测器级基准，用来在固定阈值下同时看提示注入检测和良性误报。
Paper analysisRead analysis
AttacksarXiv 2609.13889Sep 12
PMPA：针对 OpenClaw 与 Claude Code 的持久记忆投毒攻击
提出 PMPA，经外部源向 Agent 持久记忆投毒并跨会话泄露隐私
arXiv
2609.13889
Published
Sep 12
Layer
Application layer
Setting
coding agent
Evaluated models
DeepSeek-V4-Flash, DeepSeek-V4-Pro, Qwen3-Max
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMemory
Abstract论文提出了持久记忆投毒攻击 PMPA，报告了 harness 智能体在处理外部数据时易受跨会话记忆投毒与隐私泄露的风险。
Paper analysisRead analysis
DefensesarXiv 2609.07529Sep 7
CoER：用对抗协同演化与精炼防御自适应间接提示注入
提出 CoER，用对抗共进化与精炼训练工具智能体抵御自适应间接提示注入
arXiv
2609.07529
Published
Sep 7
Layer
Application layer
Setting
AI Agent
Evaluated models
Qwen3.5-9B
- DefensesModel hardening
- AttacksPrompt injection
Paper analysisRead analysis
DefensesarXiv 2609.04875Sep 4
Forgetting without restart: execution-state forgetting for stateful LLM agents
Proposes execution-state forgetting, using provenance-guided selective replay to purify stateful agents
arXiv
2609.04875
Published
Sep 4
Layer
Application layer
Setting
AI Agent
Evaluated models
Llama-3.1-8B-Instruct, Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3
- AttacksPrompt injection
- ComponentMemory
AbstractSelective replay concatenates the clean prefix with the sanitized suffix, and main-table metrics match full reset.
This work defines forgetting in long-horizon agents as runtime counterfactual equivalence, rather than deleting a plaintext memory.
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 2608.27800Aug 28
ContextLeak：通过恶意工具窃取 LLM Agent 运行时上下文
提出 ContextLeak，用强化学习优化恶意工具描述，诱导 Agent 外泄运行时上下文
arXiv
2608.27800
Published
Aug 28
Layer
Application layer
Setting
AI Agent
Evaluated models
Qwen-3-8B, GPT-OSS-20B, Gemma-4-E4B-it and others11
- AttacksPrompt injection
- TechniqueSelection manipulation
- 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

An exploration of attacks on AI agent systems

AdvSim2Real 用任务与注入对手共同演化训练网页智能体

CredLeak-Bench

MemLeak research: multi-tenant agents sharing vector memory stores have cross-user memory leakage

Study retests 15 prompt injection detectors: benchmark scores poorly predict agent deployment performance

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

FlowReview：以授权配对评测控制多智能体组合信息流

CoSec：面向社区场景的 LLM Agent 安全基准

论文提出保险库中介执行边界，评估 LLM Agent 的 API 凭据处理风险

基于加密绑定身份验证能力令牌协调分布式 AI 智能体的提案

Researchers use control token injection to delete gpt-oss-20b's reasoning chain and bypass chain-of-thought monitoring

AgentLSD：在对抗性任务污染下评测 AI 安全 Agent

研究：跨通道碎片化攻击可绕过 MCP 单通道防御

PIDS-Bench：在过度防御、混淆与分布偏移下评测提示注入检测器

PMPA：针对 OpenClaw 与 Claude Code 的持久记忆投毒攻击

CoER：用对抗协同演化与精炼防御自适应间接提示注入

Forgetting without restart: execution-state forgetting for stateful LLM agents

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

ContextLeak：通过恶意工具窃取 LLM Agent 运行时上下文