Skip to content
Search
paperOctober 2025Unreviewed

Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems

Edoardo Allegrini, Ananth Shreekumar, Z. B. Celik

arXiv.org

View resource

Abstract

Agentic AI systems, which leverage multiple autonomous agents and large language models (LLMs), are increasingly used to address complex, multi-step tasks. The safety, security, and functionality of these systems are critical, especially in high-stakes applications. However, the current ecosystem of inter-agent communication is fragmented, with protocols such as the Model Context Protocol (MCP) for tool access and the Agent-to-Agent (A2A) protocol for coordination being analyzed in isolation. Th

Categories

Agentic ThreatsAutonomous Operations

Cite

@misc{allegrini2025formalizing,
  title = {{Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems}},
  author = {Edoardo Allegrini and Ananth Shreekumar and Z. B. Celik},
  year = {2025},
  month = oct,
  eprint = {2510.14133},
  archivePrefix = {arXiv},
  doi = {10.48550/arXiv.2510.14133},
  url = {https://www.semanticscholar.org/paper/fa748fc17c5e7506bf3fed331e81b2011e5039d0}
}