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

Oct 12 · Monday

All papers

Found 10 papers

Filter4

Attack type

Defense type

Inference-time intervention2 Detection and filtering8 Instruction–data separation8 Permissions and approval13 Monitoring and auditing17

Impact

Unauthorized actions7

System type

Models and APIs47 LLM applications8 AI Agent59 coding agent6 science agent3 web agent6 Browser and computer use4 Multi-agent systems5 Embodied systems and robotics3

Attack surface / entry point

Web pages and interfaces3 Tool and skill descriptions3

Layer

Application layer10

System component

MCP and skills3 Permissions and sandboxing3

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 248papers have no tags yet and are not included in filtered results.

Defense type：Inference-time intervention Defense type：Instruction–data separation System type：coding agent System type：Browser and computer use Clear all

Another 248papers do not have category tags yet and are not included in filtered results.

Analysis and theoryarXiv 2610.04375Oct 3
Paper reveals that LLM agent tool calls are rewritten during execution paths, and proposes IntAct as a fix
提出 IEC 与 IntAct，测量并修复 Agent 工具调用的路径篡改
arXiv
2610.04375
Published
Oct 3
Layer
Application layer
Setting
coding agent
Evaluated models
qwen3-coder-plus, Qwen2.5-Coder-32B, Qwen2.5-72B and others4
- DefensesInstruction–data separation
- RiskDangerous agent actions
Abstract论文界定工具调用在执行路径中的意图偏离问题，提出无执行归因协议与跳级修复，为智能体工程提供评测与设计准则。
Paper analysisRead analysis
DefensesarXiv 2610.03089Oct 2
COBRA: A Dual-LLM Architecture for Defending Computer-Using Agents Against Branch-Jumping Attacks
提出 COBRA 双 LLM 架构，防御计算机使用 Agent 的分支转向攻击
arXiv
2610.03089
Published
Oct 2
Layer
Application layer
Setting
Browser and computer use
Evaluated models
gpt-5, UI-TARS-1.5-7B, claude-sonnet-5 and others6
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract提出分支转向防御架构 COBRA，通过事前能力约束与边界代理消除分支转向攻击并兼顾效用。
Paper analysisRead analysis
DefensesarXiv 2610.00450Oct 1
ZoneClaw 用记忆分区防御 OpenClaw 式计算机使用 Agent 的持久记忆攻击
提出 ZoneClaw，用记忆分区防御计算机使用 Agent 的持久记忆攻击
arXiv
2610.00450
Published
Oct 1
Layer
Application layer
Setting
Browser and computer use
Evaluated models
Claude Sonnet 4.6, GPT-5.5, GLM-5.2 and others4
- DefensesInstruction–data separation
- AttacksPoisoning and backdoors
- ComponentMemory
Paper analysisRead analysis
DefensesarXiv 2609.19587Sep 17
红队测试 Auto Mode：用注入与多上下文攻击绕过编码 Agent 阻断监控器
提出 Auto Mode ++，加固编码 Agent 阻断监视器
arXiv
2609.19587
Published
Sep 17
Layer
Application layer
Setting
coding agent
Evaluated models
Opus 5, Opus 4.8, Opus 4.7 and others8
- DefensesMonitoring and auditing
- RiskDeception and scheming
- ComponentMonitors
Abstract系统评估生产阻断监视器在面对持续恶意智能体时的表现，提出加固防御并指明多上下文防御挑战。
Paper analysisRead analysis
DefensesarXiv 2609.03438Sep 3
CONFLICTGUI 基准与 CONFLICTGUARD 框架：让多模态 GUI Agent 学会在冲突指令下终止执行
提出 ConflictGuard，引导 GUI Agent 在冲突指令下终止执行
arXiv
2609.03438
Published
Sep 3
Layer
Application layer
Setting
Browser and computer use
Evaluated models
UI-Venus-1.5-8B, UI-TARS-1.5-7B, OS-Atlas-Base-7B and others10
- DefensesInference-time intervention
- RiskDangerous agent actions
AbstractConflictGUI 暴露 GUI 智能体的执行偏向过度服从，ConflictGuard 用条件转向提高冲突终止率。
ConflictGuard用推理时的可行性核验和条件激活转向，让 GUI 智能体在指令不可行时终止，而不是盲执行。
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
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 2606.22673Jun 22
Agent MechSuits：面向多轮 CLI Agent 的机制子空间安全引导
提出 Agent MechSuits，对多轮 CLI Agent 做步级检测与子空间转向
arXiv
2606.22673
Published
Jun 22
Layer
Application layer
Setting
coding agent
Evaluated models
LLaMA-3.1-8B, Qwen-2.5-7B, Gemma-2-9B
- DefensesInference-time intervention
- AttacksMalicious use
- ComponentMonitors
Abstract白盒步级探针加稀疏转向可压低三模型恶意动作，同时提高生成崩溃率。
Agent MechSuits 是面向多轮 CLI 智能体的白盒、步级表示防御，并配有可执行轨迹基准 MAS。
Paper analysisRead analysis
DefensesarXiv 2605.08382May 9
SecureForge 通过提示词优化发现并减少 LLM 生成代码漏洞
提出 SecureForge，在推理时优化系统提示以减少编码智能体的代码弱点
arXiv
2605.08382
Published
May 9
Layer
Application layer
Setting
coding agent
Evaluated models
Qwen 2.5 Coder 7B, Qwen 2.5 Coder 14B, Qwen3 30B-A3B and others10
- DefensesInstruction–data separation
AbstractSecureForge 审计良性 CWE 失败并优化系统提示词。
SecureForge是一条只用 API 和静态分析、在推理时加固编码模型的流水线。
Paper analysisRead analysis
DefensesarXiv 2602.10915Feb 11
Blind Gods and Broken Screens: Building a Secure, Intent-Centric Mobile Agent Operating System
提出 Aura，用可信内核中介移动 Agent 的身份、感知与动作控制
arXiv
2602.10915
Published
Feb 11
Layer
Application layer
Setting
Browser and computer use
Evaluated models
Gemini-3-Flash
- DefensesInstruction–data separation
- AttacksPrompt injection
- ComponentPermissions and sandboxing
AbstractAura mediates structured interactions via the Agent Kernel.
Aura uses kernel-mediated structured agent interaction, replacing screen-interface mobile assistants..
Paper analysisRead analysis

You have reached the end

AI Safety Daily

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