Skip to content
Search
paperApril 2026Unreviewed

A Comparative Evaluation of AI Agent Security Guardrails

Qi Li, Jiu Li, Pingtao Wei, Jianjun Xu, Xueyi Wei, Jiwei Shi, Xuan Zhang, Yanhui Yang, Xiaodong Hui, Peng Xu, Lingquan Zhou

View resourcePDF

Abstract

This report presents a comparative evaluation of DKnownAI Guard in AI agent security scenarios, benchmarked against three competing products: AWS Bedrock Guardrails, Azure Content Safety, and Lakera Guard. Using human annotation as the ground truth, we assess each guardrail's ability to detect two categories of risks: threats to the agent itself (e.g., instruction override, indirect injection, tool abuse) and requests intended to elicit harmful content (e.g., hate speech, pornography, violence).

Categories

Prompt InjectionAgentic ThreatsGuardrailsBenchmarks & Evaluation

Framework mappings

OWASP Top 10 for LLM Applications
  • LLM01Prompt Injection
MITRE ATLAS
  • AML.T0051LLM Prompt Injection

Suggested from the entry's categories.

Cite

@misc{li2026comparative,
  title = {{A Comparative Evaluation of AI Agent Security Guardrails}},
  author = {Qi Li and Jiu Li and Pingtao Wei and Jianjun Xu and Xueyi Wei and Jiwei Shi and Xuan Zhang and Yanhui Yang and Xiaodong Hui and Peng Xu and Lingquan Zhou},
  year = {2026},
  month = apr,
  eprint = {2604.24826},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.24826}
}