Why Static Skill Inconsistencies Do Not Equate to Runtime Risks: A Behavioral Analysis of LLM Agent Execution
cs.AI
Submitted: 2026-10-13
Updated: 2026-10-13
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- GitSkills: A Dataset of Agent Skills on GitHub
- Agent Skill Evaluation and Evolution: Frameworks and Benchmarks
- Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents
- MalSkillBench: A Runtime-Verified Benchmark of Malicious Agent Skills
- SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
- Do Skill Descriptions Tell the Truth? Detecting Undisclosed Security Behaviors in Code-Backed LLM Skills
- From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills
- Cloak and Detonate: Scanner Evasion and Dynamic Detection of Agent Skill Malware
- SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
- HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?
- SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
- ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox
- SkillConsist: Detecting Inconsistencies in Agent Skills via Bidirectional Graph Alignment
- GuardianAgentBench: Where Agents Fail and How to Guard Them
- Model or Harness? An Interaction-Centric Taxonomy for Localizing Agent Failures
- SkillAlign: Aligning Skill Interfaces for LLM-based Agents
- How Do LLMs Fail In Agentic Scenarios? A Qualitative Analysis of Success and Failure Scenarios of Various LLMs in Agentic Simulations
- Description-Code Inconsistency in Real-world MCP Servers: Measurement, Detection, and Security Implications
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
- Agents' Last Exam
- 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