Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming

2026-06-30

AI Summary

Current open-source AI infrastructure lacks a unified security evaluation framework that comprehensively addresses multi-layered attack surfaces. This work proposes AI-Infra-Guard, the first open-source red-teaming framework covering four critical layers: AI agent infrastructure, protocols and tools, agent behaviors, and underlying models. It introduces an innovative “layer-paradigm alignment” mechanism, applying rule-based detection, LLM-driven auditing, black-box red-teaming, and jailbreaking assessments tailored to each layer’s threat model. The framework integrates over 1,400 vulnerability rules, deterministic checks for more than 75 components, MCP-based skill supply chain auditing, and 26+ jailbreaking attack operators. Evaluated across sixteen datasets, AI-Infra-Guard demonstrates strong effectiveness and provides the community with an extensible, full-stack foundation for AI infrastructure security.

Abstract

The fast growth of open-source AI infrastructure, from model serving engines and agent platforms to the Model Context Protocol (MCP) ecosystem and the language models themselves, has outpaced the security tooling available to defend it. We present AI-Infra-Guard, an open-source framework that organizes AI red teaming around a single observation: the attack surface of an AI agent is stratified across layers (infrastructure, protocol/tool, agent behavior, and model), and no single detection paradigm fits all of them. The framework therefore matches a paradigm to each layer, from deterministic rule matching over 75+ AI components and 1{,}400+ vulnerability rules, through LLM-driven agentic auditing of MCP servers and agent-skill packages and multi-turn black-box agent red teaming, to a jailbreak harness with 26+ attack operators over sixteen datasets. To our knowledge it is the only open-source framework to span all of these, including supply-chain auditing of the agent skills that increasingly extend AI agents. We release AI-Infra-Guard as open source so that \emph{layer-paradigm matching} can serve as a practical foundation for agent security and a shared base for the community to build on.

Problem

• AI Agent Security
• Red Teaming
• Attack Surface
• Multi-Layer Security
• AI Infrastructure

Innovation

• layer-paradigm matching
• AI red teaming
• multi-layer security
• agent supply-chain auditing
• jailbreak harness