COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol (MCP) Layers

arXiv:2610.04378 · cs.CR, cs.AI · Submitted 2026-10-03 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-10-03

Updated: 2026-10-17

Comments: Accepted as a poster at the NeurIPS 2026 Workshop on Agents in the Wild

Code: https://github.com/inspire-center/copex

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Adversarial Context — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 5 episodes

Sources

• MCPXKIT: The Unified Toolkit for Analyzing Model Context Protocol Security
• Model Context Protocol (MCP) at First Glance: Studying the Security and Maintainability of MCP Servers
• Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions
• SHADE-Arena: Evaluating Sabotage and Monitoring in LLM Agents
• MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers
• MCP-DPT: A Defense-Placement Taxonomy and Coverage Analysis for Model Context Protocol Security
• MCPTox: A Benchmark for Tool Poisoning Attack on Real-World MCP Servers
• MCPSecBench: A Systematic Security Benchmark and Playground for Testing Model Context Protocols
• InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents
• MCP Security Bench (MSB): Benchmarking Attacks Against Model Context Protocol in LLM Agents
• MCP-SafetyBench: A Benchmark for Safety Evaluation of Large Language Models with Real-World MCP Servers

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
• Estimating the Social Cost of Corporate Data Breaches
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills