Blue Teaming Function-Calling Agents

Blue Teaming Function-Calling Agents

AAAI 2026 Workshop TrustAgent Submission13 Authors

Published: 20 Nov 2025, Last Modified: 09 Mar 2026AAAI 2026 TrustAgent Workshop PosterEveryoneRevisionsBibTeXCC BY 4.0

Keywords: Agents, Function Calling, Blue Teaming

Abstract: We present an experimental evaluation that assesses the robustness of four open source LLMs claiming function-calling capabilities against three different attacks, and we measure the effectiveness of eight different defences. Our results show how these models are not safe by default, and how the defences are not yet employable in real-world scenarios.

Submission Number: 13

Loading