Metraipolis › archive › dc_c01bd16c878b
COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers — arXiv:2610.04378
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04378
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:26:10.744Z
- SHA-256 of the archived text
ce87435b006a759d6959a07db3f69693525ccf7a10d64831b63b2f0ebd5bb3de- Stable URL of this page
https://metraipolis.com/documento/dc_c01bd16c878b
How to cite
COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers — arXiv:2610.04378 — arXiv, https://arxiv.org/abs/2610.04378 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c01bd16c878b (Metraipolis archive dc_c01bd16c878b, sha256 ce87435b006a759d6959a07db3f69693525ccf7a10d64831b63b2f0ebd5bb3de)
BibTeX
@misc{dcc01bd16c878b,
title = {COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers — arXiv:2610.04378},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04378},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c01bd16c878b; sha256 ce87435b006a759d6959a07db3f69693525ccf7a10d64831b63b2f0ebd5bb3de},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04378. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. We introduce COPEX (COntext Provider EXploitation), a controlled benchmark that isolates the model as an MCP client by fixing the surrounding agent stack and varying only the tool-selecting model. COPEX covers 25 attack types instantiated as 125 scenarios across four entry surfaces: model/agent, client, server/tool, and transport. Across nine models and 3,375 trials, the mean attack success rate is 64.4%, with surface-level means ranging from 58.3% to 71.4%. Some client- and transport-level attacks succeed partly outside the model's observation or control, separating system exposure from model susceptibility. Combined input and context scanning reduces mean attack success by 49.6% on an eight-attack defense subset relative to the undefended setting. The benchmark is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.04378 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T05:24:05.843Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).