Safe Skill Retirement for Physical Agents

arXiv:2609.29543 · cs.AI · Submitted 2026-08-25 · Read on arXiv

cs.AI

Submitted: 2026-08-25

Updated: 2026-08-25

Terminology

• Benchmark — 12× in this paper · explained in 2 episodes
• Agent Skills — 11× in this paper · explained in 1 episode
• Trace — 8× in this paper · explained in 2 episodes
• Pruning — 5× in this paper · explained in 4 episodes
• Home Assistant — 4× in this paper
• Accuracy — 2× in this paper · explained in 1 episode
• Failure Modes — 2× in this paper · explained in 3 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Authentication for the Home Internet of Things — 1× in this paper
• Benchmarking — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Task Planning — 1× in this paper · explained in 2 episodes

Sources

• Defeating Prompt Injections by Design
• SkillReducer: Optimizing LLM Agent Skills for Token Efficiency
• SkillAxe: Sharpening LLM-Authored Agent Skills Through Evaluation-Guided Self-Refinement
• Gemma 4 Technical Report
• SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
• Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale
• Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
• SAGE: Smart home Agent with Grounded Execution
• Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
• Beyond OAuth: Task-Scoped Authorization for AI Agents via Natural Language Slices
• Progent: Securing AI Agents with Privilege Control
• SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection