arXiv:2610.04378v1 Announce Type: new Abstract: Large language models increasingly mediate tool use in Model Context Protocol (MCP) systems, where adversarial influence may enter through user instructions, tool schemas, tool outputs, or protocol messages. Existing benchmarks often evaluate deployed agents, conflating model susceptibility with guardrails, orchestration, and general task capability.
COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers
About this summary. This is a short, independently written summary of an article first published by arXiv cs.CR. Cyber Security News did not report or verify the underlying story. Read the original: https://arxiv.org/abs/2610.04378
Source attribution: headline and facts are from arXiv cs.CR (arxiv.org). Summary method: excerpt of the source description. See our source attribution policy.
Related coverage
INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems
arXiv:2610.20776v1 Announce Type: new Abstract: LLM-based agents increasingly execute long-horizon, end-to-end workflows involving shell commands, code modification…
GitHub adopts cheaper AI model to catch secrets developers leave in code
GitHub has trained a credential classifier to improve generic secret detection and reduce false alerts. The new pipeline cut median post-push scan execution time by 86%…
MemForm: Benchmarking Memory Formation Poisoning in LLM Agents
arXiv:2610.14431v1 Announce Type: new Abstract: Long-term memory enables LLM agents to learn from past interactions but exposes memory formation to adversarial…
APEX: Active Protection at Execution Boundaries for LLM Agents
arXiv:2610.06966v1 Announce Type: new Abstract: Indirect prompt injection (IPI) hides adversarial instructions in content that large language model (LLM) agents read at…
Understanding and Enhancing Backdoor Persistency in LLM Agent Post-Training
arXiv:2610.07510v1 Announce Type: new Abstract: Developers can build LLM agents by adapting third-party models through benign post-training. We study a supply-chain…
HarnessSecurity-Bench: Do Security Mechanisms Really Protect Coding Agent Harnesses?
arXiv:2610.07639v1 Announce Type: new Abstract: Coding agent harnesses mediate tool use and authorize actions, yet their security mechanisms and runtime effects remain…