paper-with-me

Papers

A Comparative Evaluation of AI Agent Security Guardrails

2026-04-27 · Qi Li, Jiu Li, Pingtao Wei, Jianjun Xu, Xueyi Wei, Jiwei Shi, Xuan Zhang, Yanhui Yang, Xiaodong Hui, Peng Xu, Lingquan Zhou 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.

📄 PDF Abstract BibTeX arXiv:2604.24826

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Don't Make Models Guess Security and Safety: Symbolic Guardrails for Domain-Specific AI Agents

2026-04-16 · Yining Hong, Yining She, Eunsuk Kang, Christopher S. Timperley 외 arxiv

There is increasing interest in integrating AI agents that invoke tools into domain-specific commercial software, where unintended tool calls can cause serious security and safety incidents. This has drawn growing resear…

LlamaFirewall: An open source guardrail system for building secure AI agents

2025-05-06 · Sahana Chennabasappa, Cyrus Nikolaidis, Daniel Song, David Molnar 외

Large language models (LLMs) have evolved from simple chatbots into autonomous agents capable of performing complex tasks such as editing production code, orchestrating workflows, and taking higher-stakes actions based o…

Chatbot

CS-Guard: Benchmarking LLM Guardrails for Code Generation Security

2026-09-09 · Jinyang Li, Mingyu Guo, Hung X. Nguyen arxiv

Large language models (LLMs) have been ex- ploited to generate malware, but the effective- ness of guardrails for code generation secu- rity remains unclear. We introduce CS-Guard, the first benchmark to systematically e…

Text-to-Code GenerationCode TranslationCode Completion

SkillTester: Benchmarking Utility and Security of Agent Skills

2026-03-28 · Leye Wang, Zixing Wang, Anjie Xu arxiv

This technical report presents SkillTester, a tool for evaluating the utility and security of agent skills. Its evaluation framework combines paired baseline and with-skill execution conditions with a separate security p…

No Free Lunch with Guardrails

2025-04-01 · Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa, Sahil Agarwal 외

As large language models (LLMs) and generative AI become widely adopted, guardrails have emerged as a key tool to ensure their safe use. However, adding guardrails isn't without tradeoffs; stronger security measures can …