展示HN:Revenant – 基于自动化大型语言模型的逆向工程与重实现工具

3作者: sylwester大约 24 小时前原帖
我是一名硬件工程师和安全研究员,一直在思考我的工作是否可以部分自动化,以便我能够专注于其他主题。因此,我开发了revenant——一个基于大型语言模型(如Claude、OpenAI和本地AI)的工具包,围绕radare2、ghidra等构建,能够完全自动化地分析固件,实施开源框架,包括引脚分配、硬件启动、外设启动等,甚至可以1:1复制现有固件,从而使旧硬件能够使用现代工具链复活。 一些应用场景包括: - 赋予旧硬件新生命 - 对可疑固件进行安全分析 请查看: [https://github.com/DatanoiseTV/revenant](https://github.com/DatanoiseTV/revenant) 我非常希望能得到一些反馈,也许还有一些改进建议。
查看原文
I am a hardware engineer and security researcher and I&#x27;ve been wondering whether my work could be partially automated, so I can focus on other topics as well, so I build revenant - a LLM powered (Claude, OpenAI, local AI) toolkit that builds around radare2, ghidra etc and can fully automatically analyze firmware, implement open source skeletons incl. pinouts, hardware bringup, peripheral bringup etc. or can even 1:1 replicate existing firmware so old hardware can be resurrected with modern toolchains.<p>Some applications are:<p>- Give old hardware new life<p>- Security Analysis of shady firmware<p>Check it at: <a href="https:&#x2F;&#x2F;github.com&#x2F;DatanoiseTV&#x2F;revenant" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;DatanoiseTV&#x2F;revenant</a><p>I would love some input on this and maybe some recommendations for improvements.