Image de couverture de livre

The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Fan, Linfeng, Li, Ziwei, Tian, Yuan, Wang, Yichen, Li, Rongsheng, Wang, Xiong
Format:	Preprint
Publié:	2026
Sujets:	Cryptography and Security Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Reframing LLM Agent Security as an Agent-Human Interaction Problem
par: Wang, Peiran, et autres
Publié: (2026)

Granite: Granular Runtime Enforcement for GitHub Actions Permissions
par: Moazen, Mojtaba, et autres
Publié: (2025)

DRIFT: Dynamic Rule-Based Defense with Injection Isolation for Securing LLM Agents
par: Li, Hao, et autres
Publié: (2025)

SoK: Trust-Authorization Mismatch in LLM Agent Interactions
par: Shi, Guanquan, et autres
Publié: (2025)

SkillScope: Toward Fine-Grained Least-Privilege Enforcement for Agent Skills
par: Wu, Jiangrong, et autres
Publié: (2026)

Agent-Sentry: Bounding LLM Agents via Execution Provenance
par: Sequeira, Rohan, et autres
Publié: (2026)

ALPS: Automated Least-Privilege Enforcement for Securing Serverless Functions
par: Shin, Changhee, et autres
Publié: (2026)

Secure and Privacy-Preserving Authentication for Data Subject Rights Enforcement
par: Hansen, Malte, et autres
Publié: (2024)

MemLineage: Lineage-Guided Enforcement for LLM Agent Memory
par: Ouyang, Ciyan, et autres
Publié: (2026)

KnowHow: Automatically Applying High-Level CTI Knowledge for Interpretable and Accurate Provenance Analysis
par: Meng, Yuhan, et autres
Publié: (2025)

TRUCE: TRUsted Compliance Enforcement Service for Secure Health Data Exchange
par: Kim, Dae-young, et autres
Publié: (2025)

NodeShield: Runtime Enforcement of Security-Enhanced SBOMs for Node.js
par: Cornelissen, Eric, et autres
Publié: (2025)

Securing LLM Agents Need Intent-to-Execution Integrity
par: Qu, Wenjie, et autres
Publié: (2026)

Maris: A Formally Verifiable Privacy Policy Enforcement Paradigm for Multi-Agent Collaboration Systems
par: Cui, Jian, et autres
Publié: (2025)

"Security vs. Interoperability" Arguments: An Analytical Framework
par: Landis, Daji, et autres
Publié: (2025)

SecTracer: A Framework for Uncovering the Root Causes of Network Intrusions via Security Provenance
par: Lee, Seunghyeon, et autres
Publié: (2025)

A Survey of LLM-Driven AI Agent Communication: Protocols, Security Risks, and Defense Countermeasures
par: Kong, Dezhang, et autres
Publié: (2025)

RulePilot: An LLM-Powered Agent for Security Rule Generation
par: Wang, Hongtai, et autres
Publié: (2025)

MindGuard: Intrinsic Decision Inspection for Securing LLM Agents Against Metadata Poisoning
par: Wang, Zhiqiang, et autres
Publié: (2025)

LLM-driven Provenance Forensics for Threat Investigation and Detection
par: Mukherjee, Kunal, et autres
Publié: (2025)

UC-Secure Star DKG for Non-Exportable Key Shares with VSS-Free Enforcement
par: Sehrawat, Vipin Singh
Publié: (2026)

Constraining Host-Level Abuse in Self-Hosted Computer-Use Agents via TEE-Backed Isolation
par: Lu, Di, et autres
Publié: (2026)

The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
par: Wang, Peiran, et autres
Publié: (2026)

RAS-Eval: A Comprehensive Benchmark for Security Evaluation of LLM Agents in Real-World Environments
par: Fu, Yuchuan, et autres
Publié: (2025)

IPBAC: Interaction Provenance-Based Access Control for Secure and Privacy-Aware Systems
par: Zisad, Sharif Noor, et autres
Publié: (2026)

Security Approaches for Data Provenance in the Internet of Things: A Systematic Literature Review
par: Faraj, Omair, et autres
Publié: (2024)

A Framework for Formalizing LLM Agent Security
par: Siu, Vincent, et autres
Publié: (2026)

SoK: Security of Autonomous LLM Agents in Agentic Commerce
par: Mao, Qian'ang, et autres
Publié: (2026)

Incorporating Gradients to Rules: Towards Lightweight, Adaptive Provenance-based Intrusion Detection
par: Wang, Lingzhi, et autres
Publié: (2024)

VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability Detection
par: Nie, Yuzhou, et autres
Publié: (2025)

Picachv: Formally Verified Data Use Policy Enforcement for Secure Data Analytics
par: Chen, Haobin Hiroki, et autres
Publié: (2025)

OpenClaw PRISM: A Zero-Fork, Defense-in-Depth Runtime Security Layer for Tool-Augmented LLM Agents
par: Li, Frank
Publié: (2026)

MPC-EVM: Enabling MPC Execution by Smart Contracts In An Asynchronous Manner
par: Zhou, Yichen, et autres
Publié: (2025)

ChainCaps: Composition-Safe Tool-Using Agents via Monotonic Capability Attenuation
par: Jiang, Xiaochong, et autres
Publié: (2026)

PentestAgent: Incorporating LLM Agents to Automated Penetration Testing
par: Shen, Xiangmin, et autres
Publié: (2024)

LLM Agents Should Employ Security Principles
par: Zhang, Kaiyuan, et autres
Publié: (2025)

Marlin: Knowledge-Driven Analysis of Provenance Graphs for Efficient and Robust Detection of Cyber Attacks
par: Li, Zhenyuan, et autres
Publié: (2024)

TraceAegis: Securing LLM-Based Agents via Hierarchical and Behavioral Anomaly Detection
par: Liu, Jiahao, et autres
Publié: (2025)

Reasoning-Style Poisoning of LLM Agents via Stealthy Style Transfer: Process-Level Attacks and Runtime Monitoring in RSV Space
par: Zhou, Xingfu, et autres
Publié: (2025)

Defensible Design for OpenClaw: Securing Autonomous Tool-Invoking Agents
par: Li, Zongwei, et autres
Publié: (2026)