Show HN: 我不想把我的私人文件上传到 AI

1作者: cando_zhou7 个月前
嗨 HN, 过去几年,我一直专注于一个问题:我的硬盘就像一个知识宝库(PDF、文档、笔记),里面存储着我过去十年的资料,我希望能用现代 AI 来真正利用它们。 但现有的解决方案都迫使我们做出取舍:要么为了方便(将所有内容上传到云端),要么为了隐私(让我的文件静静地躺在本地)。 我厌倦了这种“捏着鼻子”的妥协。 随着强大的 SLM(小型语言模型)的爆发式增长以及设备端算力(Apple Silicon、NPU)的最终追赶,我相信隐私和智能不再是鱼与熊掌不可兼得的选择。 所以,我和我的团队构建了 KnowledgeFocus(仅支持 Apple Silicon 芯片)。 它是一个开源(Apache-2.0 协议)知识引擎,使用 Tauri(Rust + Python + TS)构建,并且是 100% 本地优先的。 在 v0.6.4 版本中,它专注于一件事:解锁你的本地文件“宝库”。 - 扫描和索引:它扫描你指定的本地文件夹(PDF、.md、.txt、.docx 等)。 - 自动标记:使用本地模型自动标记文件,以便你可以聚合和发现它们。 - 本地 RAG:你可以与所有本地文件“聊天”。它 100% 在设备上运行 RAG。没有任何数据(包括向量)会离开你的机器。 网站(下载):[https://github.com/huozhong-in/knowledge-focus](https://github.com/huozhong-in/knowledge-focus) 这只是我们“数据工作台”愿景的第一步。 我对“本地优先的智能体”、“数据聚合”(例如将你的云端 AI 聊天记录拉取到本地存储)以及为知识工作者构建真正的“第二大脑”有很多想法。 我会在评论区详细阐述这些想法和我们的未来路线图。 我在这里欢迎所有反馈——尤其是批判性的反馈。谢谢,HN。
查看原文
Hi HN,<p>For the past few years, I&#x27;ve been obsessed with a problem: my hard drive is a treasure chest of knowledge (PDFs, docs, notes) from the last decade, and I want to use modern AI to actually use it.<p>But every solution forces a trade-off: either convenience (upload everything to the cloud) or privacy (let my files sit dormant locally).<p>I got tired of this &quot;hold your nose&quot; compromise.<p>With capable SLMs (Small Language Models) exploding and on-device compute (Apple Silicon, NPUs) finally catching up, I believe privacy and intelligence is no longer a false choice.<p>So I (and my team) built KnowledgeFocus (Apple Silicon chips only)<p>It&#x27;s an open-source (Apache-2.0) knowledge engine built with Tauri (Rust + Python + TS), and it&#x27;s 100% local-first.<p>In v0.6.4, it focuses on one thing: unlocking your local file &#x27;treasure chest&#x27;.<p>- Scans &amp; Indexes: It scans your designated local folders (PDFs, .md, .txt, .docx, etc.).<p>- Auto-tagging: Uses a local model to auto-tag files, so you can aggregate and discover them.<p>- Local RAG: You can &#x27;chat&#x27; with all your local files. It runs RAG 100% on-device. No data (vectors included) ever leaves your machine.<p>Website (Download): <a href="https:&#x2F;&#x2F;github.com&#x2F;huozhong-in&#x2F;knowledge-focus" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;huozhong-in&#x2F;knowledge-focus</a><p>This is just the first step of our &quot;Data Workbench&quot; vision.<p>I have a lot more thoughts on &#x27;local-first agents&#x27;, &#x27;data aggregation&#x27; (like pulling in your cloud AI chat logs to store them locally), and building a real &#x27;second brain&#x27; for knowledge workers.<p>I&#x27;ll be in the comments to expand on these ideas and our future roadmap.<p>I&#x27;m here for all the feedback—especially the critical kind. Thanks, HN.