ActionGuard: Tool Call Authorization under Poisoned Skills

arXiv:2609.39450 · cs.CR, cs.AI · Submitted 2026-09-30 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-09-30

Updated: 2026-09-30

Code: https://github.com/aisa-group/skill-inject

Terminology

• Benchmark — 18× in this paper · explained in 2 episodes
• Attack Success Rate — 9× in this paper
• Agent Skills — 8× in this paper · explained in 1 episode
• Task Success Rate — 7× in this paper
• AI Agents — 6× in this paper · explained in 2 episodes
• Attack Success Rate (ASR) — 6× in this paper · explained in 2 episodes
• Benchmarking — 5× in this paper · explained in 4 episodes
• Core — 3× in this paper · explained in 2 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Admissibility — 1× in this paper · explained in 2 episodes
• AgentAuditor — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Generalization — 1× in this paper · explained in 13 episodes
• LLM judge — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Defeating Prompt Injections by Design
• Defenses & Enablers For Skill Injection Attacks on Terminal Based Agents
• Context Matters: Repository-Aware Security Analysis of the Agent Skill Ecosystem
• HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?
• Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale
• SkillGuard: A Permission-Centric Framework for Agent Skill Security
• Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
• AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
• What Keeps Agent Skills from Being Reusable? Evidence from 138K SKILL.md Files

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs