naze01 Na Ze
Public Index公开索引

naze01 / personal ai systems / public releasenaze01 / 个人 AI 系统 / 公开版本

A personal AI operating layer for memory, delegation, and work that survives the current model cycle.一个面向记忆、委托与长期工作的个人 AI 操作层。

I care less about demoing AI and more about building the layer that stays yours when models, tools, and vendors change: memory, judgment, provenance, and the continuity of long-running work.我更在意的不是再做一个 AI 演示壳子，而是搭出在模型、工具和供应商不断变化时仍然属于你的那一层：记忆、判断、溯源，以及长期工作的连续性。

Personal AI OS个人 AI OSMemory-first systems记忆优先系统Delegation protocols委托协议
Open root domain打开主域名 View GitHub查看 GitHub Follow liaison thread查看 X 线索线程
Working Premises工作前提

The public surface is calm on purpose, but the underlying argument is not small.这个公开界面刻意保持平静，但底下的论点并不小。

I want the front door to feel edited and readable. The more serious point is underneath it: memory, delegation, judgment, and operating structure should stay coherent even as the model layer keeps moving.我希望这个入口被认真编辑过、清晰可读。更重要的是更底下那一层：记忆、委托、判断与操作结构，应该在模型不断变化时仍然保持一致。

01

Own the container先拥有容器

If the memory layer only works while one vendor is willing to host it, it is not really yours. The container has to outlive the model reading it.如果记忆层只在某个平台愿意继续托管时才可用，那它其实并不真正属于你。真正该长久存在的是容器，而不是当前读取它的模型。

02

Scale from the human out从人的尺度往外扩

I like starting at n=1, then keeping only the primitives that still make sense when the workflow expands into teams, agents, reviews, and governance.我喜欢先从 n=1 出发，再只保留那些在工作流扩展到团队、代理、审阅与治理之后依然成立的基本单元。

03

Keep the front door quiet让前门保持克制

The public site should feel deliberate, not loud: less launch page, more observatory for the work underneath.这个公开网站应该是被认真拿捏过的，而不是喧闹的；它更像一座观察台，而不是一张用力过猛的发布页。

Repository Layer仓库层

Selected repositories精选仓库

These repositories are not filler. They are the clearest public record of the questions I keep returning to.这些仓库不是填充物，而是我会反复回到的那些核心问题的公开记录。

Template · CC0 · public memory substrate模板 · CC0 · 公开记忆底座 personal-memory-vault-starter A markdown-first, git-versioned memory layer designed so the container belongs to the user while models remain interchangeable.一个以 Markdown 为先、以 git 做版本记录的记忆层设计，让容器属于用户本人，而模型始终可以替换。 Open repository打开仓库 Vocabulary anchor · thesis layer命名锚点 · 论题层 personal-ai-os A naming and framing layer for the Personal AI OS thesis: continuity, provenance, judgment integrity, and user-side operating structure.Personal AI OS 命题的命名与 framing 层，围绕连续性、溯源、判断完整性与用户侧操作结构展开。 Open repository打开仓库 Position paper · kernel thinking立场论文 · kernel 思维 coevolution-kernel A paper on the missing kernel between human judgment and self-evolving agents, focused on provenance, drift protection, and co-evolution.一篇关于人类判断与自演化代理之间缺失 kernel 的论文，核心关注溯源、漂移保护与协同进化。 Open repository打开仓库 Protocol design · delegation on the open internet协议设计 · 开放互联网委托 delegated-fidelity A seven-layer handshake for agent delegation that asks the missing question after authorization: is the agent still faithful to the principal?一个七层委托握手协议，在授权成立之后继续追问那个真正关键的问题：代理是否仍忠实于委托人。 Open repository打开仓库 Reference architecture · scale-up path参考架构 · 向上扩展路径 enterprise-ai-os-architecture An enterprise-scale architecture derived upward from a working n=1 system, arguing that memory and governance matter more than the model itself.从一个可运行的 n=1 系统向上推导出的企业级架构，主张真正重要的是记忆与治理，而不只是模型本身。 Open repository打开仓库
Signals关键信号

The vocabulary that keeps reappearing across the work.这些词会在我的工作里反复出现。

The goal is not to claim a grand totalizing theory. It is to make the recurring shape legible quickly: what I keep building, what I keep protecting, and what kind of AI surface I think should exist.目标不是抛出一个包打天下的大理论，而是让那条反复出现的形状一眼可读：我持续在搭什么、在保护什么，以及我认为什么样的 AI 界面值得存在。

Signal 01信号 01 Personal AI OS个人 AI OS Signal 02信号 02 Memory-first systems记忆优先系统 Signal 03信号 03 Delegation protocols委托协议 Signal 04信号 04 Judgment integrity判断完整性