Oxhumanode/Warpcast

BrowseSafe-Bench是一个全面的安全基准测试，旨在评估AI浏览器代理在现实HTML环境中对提示注入攻击的鲁棒性。与之前专注于简单文本注入的基准测试不同，BrowseSafe-Bench强调环境真实性，包含复杂的HTML结构、多样化的攻击语义和模仿真实世界网络复杂性的良性“干扰”元素。该数据集支持论文《BrowseSafe: Understanding and Preventing Prompt Injection Within AI Browser Agents》的研究，既是防御机制的训练场，也是前沿LLM的严格评估套件。

BrowseSafe-Bench is a comprehensive security benchmark designed to evaluate the robustness of AI browser agents against prompt injection attacks embedded in realistic HTML environments. Unlike prior benchmarks that focus on simple text injections, BrowseSafe-Bench emphasizes environmental realism, incorporating complex HTML structures, diverse attack semantics, and benign distractor elements that mimic real-world web complexity. This dataset was developed to support the research presented in the paper *BrowseSafe: Understanding and Preventing Prompt Injection Within AI Browser Agents*. It is both a training ground for defense mechanisms and a rigorous evaluation suite for frontier LLMs.