Incalmo：一种自主的LLM辅助系统用于多主机网络红队测试
Incalmo: An Autonomous LLM-assisted System for Red Teaming Multi-Host Networks
AI总结：
Incalmo是一种基于LLM的自主红队测试系统，通过高层次任务规划和专用任务代理实现多主机网络攻击，显著提高了红队测试的效率和成功率。
摘要：
安全操作员使用红队来模拟真实攻击者并主动发现防御漏洞。在现实的企业环境中，这涉及执行跨越多个“中继”主机的多主机网络攻击。不幸的是，红队测试成本高昂且需要大量专业知识和努力。鉴于LLM在CTF挑战中的潜力，我们首先分析LLM是否能够自主执行多主机红队测试。我们发现，最先进的LLM辅助攻击系统（例如PentestGPT、CyberSecEval3）配合领先的LLM（例如Sonnet 4、Gemini 2.5 Pro）无法完成此类任务。基于我们对现有系统失败模式的理解，我们认为需要改进LLM辅助红队测试的抽象和接口。基于这一见解，我们提出了Incalmo的设计和实现，这是一种用于自主多主机网络红队测试的LLM辅助系统。Incalmo利用LLM来规划红队测试，以高层次的声明性任务执行由领域特定任务代理执行。Incalmo还利用辅助服务来管理上下文和获取的资产。为了评估，我们开发了MHBench，一种新的多主机攻击基准，包含40个真实的模拟网络（从22到50个主机）。我们发现，Incalmo在37个MHBench环境中成功获取了关键资产（即关键主机或数据）。相比之下，最先进的LLM辅助系统仅在40个环境中的3个成功。我们显示，Incalmo在攻击效率方面表现出色，成功攻击耗时12-54分钟，成本低于15美元的LLM信用。
英文摘要：
Security operators use red teams to simulate real attackers and proactively find defense gaps. In realistic enterprise settings, this involves executing multi-host network attacks spanning many "stepping stone" hosts. Unfortunately, red teams are expensive and entail significant expertise and effort. Given the promise of LLMs in CTF challenges, we first analyze if LLMs can autonomously execute multi-host red team exercises. We find that state-of-the-art LLM-assisted offense systems (e.g., PentestGPT, CyberSecEval3) with leading LLMs (e.g., Sonnet 4, Gemini 2.5 Pro) are unable to do so. Building on our observations in understanding the failure modes of state-of-the-art systems, we argue the need to improve the abstractions and interfaces for LLM-assisted red teaming. Based on this insight, we present the design and implementation of Incalmo, an LLM-assisted system for autonomously red teaming multi-host networks. Incalmo uses LLMs to plan red team exercises in terms of high-level declarative tasks that are executed by domain-specific task agents. Incalmo also uses auxiliary services to manage context and acquired assets. For our evaluation, we develop MHBench, a novel multi-host attack benchmark with 40 realistic emulated networks (from 22 to 50 hosts). We find that Incalmo successfully acquires critical assets (i.e., key hosts or data) in 37 out of 40 MHBench environments. In contrast, state-of-the-art LLM-assisted systems succeed in only 3 out of 40 environments. We show that Incalmo is efficient-successful attacks took 12-54 minutes and cost <$15 in LLM credits.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
