# A Comparative Evaluation of AI Agent Security Guardrails

arXiv: https://arxiv.org/abs/2604.24826 · HTML: https://mlsanity.com/paper/2604.24826/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Prompt injection and agent security, one of 12 directions in LLM safety and security.
- Builds on 6 of its 10 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Placed in

- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 6 of 6)

- [SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2402.05044/) (2024)
- [Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game](https://mlsanity.com/paper/2311.01011/) (2023)
- [Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment](https://mlsanity.com/paper/2402.19085/) (2024)
- [ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming](https://mlsanity.com/paper/2404.08676/) (2024)
- [CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns](https://mlsanity.com/paper/2601.00588/) (2026)
- [DeepKnown-Guard: A Proprietary Model-Based Safety Response Framework for AI Agents](https://mlsanity.com/paper/2511.03138/) (2025)

## Built on by (most cited first, 1 of 1)

- [Registry-Governed Agent Lifecycle:Completing EDDOps with Evaluation-DrivenRegistration, Promotion, and Retirement on AWS AgentCore](https://mlsanity.com/paper/2607.00345/) (2026)

## Abstract (arXiv)

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). Evaluation results demonstrate that DKnownAI Guard achieves the highest recall rate at 96.5% and ranks first in true negative rate (TNR) at 90.4%, delivering the best overall performance among all evaluated guardrails.