Great AI tools,
right in your hands.把好用的 AI，
直接放到你手里。

OpenAI Disclosure: Another Australian Government Agency Hacked by a Runaway AgentOpenAI 披露：澳大利亚又一政府机构遭失控智能体入侵

After Trump proposed to rename “artificial intelligence” as “superintelligence”, Slovenia's “.si” domain name registration increased by 2100% month-on-month特朗普提出将“人工智能”改称“超级智能”后，斯洛文尼亚“.si”域名注册量环比暴增 2100%

Microsoft CEO Nadella reiterates Copilot's AI positioning: a new work-oriented operating system微软 CEO 纳德拉重申 Copilot AI 定位：面向工作的新操作系统

Another head of security, David Robinson, leaves at a time of heightened unrest within OpenAIOpenAI 内部动荡加剧之际，又一名安全负责人戴维 · 罗宾逊离职

IT Morning Post 1003: Samsung raises S26 prices; Qianjie M6 launches 3-year 0% car purchase plan; Anthropic goes on sale as early as mid-November; Apple expects to release MacBook with touch OLED screen this month...IT早报 1003：三星上调国行 S26 售价；问界 M6 推 3 年 0 息购车方案；曝 Anthropic 最早 11 月中旬上市；苹果预计本月发布触控 OLED 屏 MacBook...

Apple tightens full disk access such as macOS 27 to reduce the risk of AI-powered gymnastics苹果收紧 macOS 27 等完全磁盘访问权限，降低 AI 智能体操控风险

Apple's homeOS foresight: integrating iOS/iPadOS/watchOS features, featuring Siri AI interaction苹果 homeOS 前瞻：整合 iOS / iPadOS / watchOS 特性，主打 Siri AI 交互

Anthropic plans to spend $100 million to train 10,000 engineers to crack the "talent shortage" of enterprise AIAnthropic 拟斥资 1 亿美元培训万名工程师，破解企业 AI 落地“人才荒”

Teletext tutorial: Five minutes to build a no-vector rag with PageIndex that costs nothing图文教程：五分钟用 PageIndex 搭一个不花一分钱的无向量 RAG

Teach you how to use the PageIndex warehoused today: no vector database, no Embedding service, use the free quota LLM to build a tree-like index for your PDF, and then connect any large model for question and answer. Attached are diagrams of official benchmark data.

手把手教你用今天入库的 PageIndex：不用向量数据库、不买 Embedding 服务，拿免费额度的 LLM 给自己的 PDF 建树状索引，再接上任何大模型做问答。附官方基准数据图解。

Graphic tutorial: 3 steps to install Coucou, Claude Code permission request one-click approval图文教程：三步装好 Coucou，Claude Code 权限请求一键批准

Hands-on instructions for installing today's Coucou: From downloading and unzipping, authorization to accessing Claude Code hooks, click through the screenshot steps, and let the screen bang for 5 minutes to stare at a small creature for your conversation.

手把手教你安装今天入库的 Coucou：从下载解压、授权到接入 Claude Code hooks，全部点击级截图步骤，5 分钟让屏幕刘海多一只帮你盯会话的小生物。

Tutorial: Four steps to build an agent sandbox with strategy教程：四步给智能体建一个带策略的沙箱

4-Step Running Pass: Confirm the platform, install the CLI and local gateway, build the first sandbox, run a real agent with OpenCode, and experience the process of "approving new permissions". Works with today's NVIDIA OpenShell clients.

四步跑通：确认平台、一行装好 CLI 与本地网关、建第一个沙箱、再用 OpenCode 跑一个真实智能体并体验「批准新权限」的流程。配合今天的 NVIDIA OpenShell 客户端使用。

Tutorial: Render your first MP4 in 4 steps with HTML教程：四步用 HTML 渲染出第一条 MP4

4-Step Running Pass: Confirm the environment dependency, initialize the project, preview in the browser in real time, and render to MP4. Add another step to hand over the skills to your agent, and the verbal description of the needs will be released in the future. Works with today's HyperFrames plugin packages.

四步跑通：确认环境依赖、初始化项目、在浏览器里实时预览、渲染成 MP4。再补一步把技能交给你的智能体，以后口头描述需求就能出片。配合今天的 HyperFrames 插件包使用。

Coucou: Claude Code companion who lives in the bangs of the screen (v0.1.2 official installation package)Coucou：住在屏幕刘海里的 Claude Code 小伴侣（v0.1.2 官方安装包）

Top 8 gadgets on the GitHub daily list: a small animated creature that lives in your Mac Liu Hai (Windows is at the top of the screen), staring at every Claude Code session in real time - it flips over when the agent is working, it raises its hand when you approve the permission request, and releases or rejects it with a tap. 2982 Star, open source free, purely local operation.

登上 GitHub 日榜前八的小工具：一只住在你 Mac 刘海（Windows 则在屏幕顶部）里的动画小生物，实时盯着每一个 Claude Code 会话——智能体干活时它翻跟头，等你批准权限请求时它举手示意，点一下就放行或拒绝。2982 星、开源免费、纯本地运行。

HyperFrames v0.8.107: Writing videos in HTML, HeyGen's official rendering plugin for agentsHyperFrames v0.8.107：用 HTML 写视频，HeyGen 官方给智能体的渲染插件

On the same day, it was on the GitHub Hot List, an open source project of 55,000 Stars. The idea is very clean: the video is no longer dragged on the timeline, but written as an HTML-CSS animation, GSAP timeline, media materials are put in, and finally definitively rendered into MP4. The agent-oriented plug-in package (including 21 skills) was officially released with v0.8.107, about 11.87 MB.

当天登上 GitHub 热榜、5.5 万 Star 的开源项目。思路很干净：视频不再用时间轴拖拽，而是写成一份 HTML——CSS 动画、GSAP 时间轴、媒体素材都放进去，最后确定性渲染成 MP4。官方随 v0.8.107 发布了面向智能体的插件包（含 21 个技能），约 11.87MB。

OpenRig v0.6.2: Turn Claude Code and Codex into one agent squad with a single YAML fileOpenRig v0.6.2：用一份 YAML 把 Claude Code 和 Codex 编成一支智能体小队

A GitHub Trending project that shipped v0.6.2 today, around 2,700 stars. The idea is plain: a harness wraps one model, a rig wraps all of your harnesses. Define the team in YAML, boot it with one command, and run Claude Code plus Codex side by side inside the same rig. Official npm offline package, about 5.95 MB.

当天发布 v0.6.2 的 GitHub 热榜项目，约 2700 Star。它的思路很直白：harness 包的是一个模型，rig 包的是你所有的 harness——把队伍写进 YAML，一条命令启动，Claude Code 与 Codex 同处一个 rig 里按图编队。官方 npm 离线包约 5.95MB。

OpenAI official Codex plug-in is installed into Claude Code: commanding two programming agents at the same time in one terminalOpenAI 官方 Codex 插件装进 Claude Code：一个终端里同时指挥两个编程智能体

GitHub hot list project this week, about 34,000 stars. It directly connects Codex to Claude Code: you can let Claude do the work, let Codex review it independently, or you can dump a stuck task in its entirety to the Codex background for running. The plug-in itself is only about 118 KB and can be installed with three commands from Claude Code’s plug-in market.

本周 GitHub 热榜项目，约 3.4 万 Star。它把 Codex 直接接进 Claude Code：你可以让 Claude 干活、让 Codex 独立复核，也能把一个卡住的任务整包甩给 Codex 后台跑。插件本体仅约 118 KB，靠 Claude Code 的插件市场三条命令装好。

PageIndex v0.2.21: Do not index “vectorless rag” documents for vector databasesPageIndex v0.2.21：不要向量数据库的「无向量 RAG」文档索引

VectifyAI open source, 38,000 Star rag new idea: do not cut the document, do not do Embedding, do not build a vector library, but let LLM directly generate a tree-like directory index for the document, retrieved like a human flip book according to directory reasoning positioning, accuracy in FinanceBench and other benchmarks inverse traditional vector rag, indexing cost is also one order of magnitude cheaper.

VectifyAI 开源、3.8 万 Star 的 RAG 新思路：不给文档切块、不做 Embedding、不建向量库，而是让 LLM 直接给文档生成一棵树状目录索引，检索时像人翻书一样按目录推理定位，准确率在 FinanceBench 等基准上反超传统向量 RAG，索引成本还便宜一个量级。

NVIDIA OpenShell v0.1.2: Give the agent a "locked room", the official Linux version is about 9.96 MBNVIDIA OpenShell v0.1.2：给智能体一间「上锁的房间」，官方 Linux 版约 9.96MB

When Nvidia's open source agents are running safely, GitHub's hot list is the first echelon on the day. It solves a very real problem: the more capable the agent is, the more it needs to read files, load dependencies, call APIs, use credentials - and these permissions are difficult to take back once they are released. OpenShell allows agents to retain these capabilities without giving them unlimited access to your data, keys, and network.

英伟达开源的智能体安全运行时，当天 GitHub 热榜第一梯队。它要解决的是一个很现实的问题：智能体越能干，就越需要读文件、装依赖、调 API、用凭据——而这些权限一旦放开就很难收回。OpenShell 让智能体保有这些能力，同时不给它对你数据、密钥和网络的无限访问权。

Hindsight v0.10.2: Long-term memory that lets your agent actually learnHindsight v0.10.2：给智能体一套会「学」的长期记忆

One of the fastest-rising memory systems today, around 43,500 stars. It stores facts, entity relationships and temporal reasoning in PostgreSQL plus pgvector, so an agent remembers people, events and time, and forms opinions under configurable disposition traits. One docker run starts the service, and an MCP endpoint comes built in.

当天涨星最猛的记忆系统之一，约 4.35 万 Star。它把事实、实体关系和时序推理存进 PostgreSQL + pgvector：智能体能记住人、事、时间，还能按可配置的性格倾向形成判断。一条 docker run 起服务，自带 MCP 端点。

Tencent Cloud open source Octop 1.0.2b4: a multi-user, multi-agent self-hosted AI assistant installed in one process腾讯云开源 Octop 1.0.2b4：一个进程装下多用户、多智能体的自托管 AI 助手

This week's GitHub hot list project, about 5.5k stars. A single process startup provides a web console, command line, eight major IM channels (Feishu, DingTalk, QQ, WeChat, Telegram, Discord, Enterprise WeChat, etc.) and natural language scheduled tasks. All data falls in the ~/.octop/ directory of the local machine.

本周 GitHub 热榜项目，约 5.5k Star。单进程启动就同时提供 Web 控制台、命令行、八大 IM 渠道（飞书、钉钉、QQ、微信、Telegram、Discord、企业微信等）与自然语言定时任务，所有数据落在本机 ~/.octop/ 目录里。

Google's Official Agent Skills Collection: Equipping Agents with Google Family Bucket SkillsGoogle 官方 Agent Skills 合集：给智能体装上 Google 全家桶技能

Google's official open source agent skills warehouse (20,000 + stars): wraps Google product capabilities such as search, YouTube, Google Sheets, and Maps into standard Agent Skill, and 77 + clients such as Claude Code and Gemini CLI are directly installed and used. This site has packaged the latest version of all skill files in October, with Chinese installation instructions.

Google 官方开源的智能体技能仓库（2 万+ Star）：把搜索、YouTube、Google 表格、地图等 Google 产品能力封装成标准 Agent Skill，Claude Code、Gemini CLI 等 77+ 客户端直接安装即用。本站已打包 10 月最新版全部技能文件，附中文安装说明。

Ponytail v4.10.0: Let AI write code like the laziest senior engineer, write 54% less in practicePonytail v4.10.0：让 AI 像最懒的资深工程师那样写代码，实测少写 54%

150,000 Stars, a regular skill pack that is still rising on the day. It's an interesting setup - stuff your AI with the kind of senior engineer who “has a long ponytail, elliptical glasses, and starts before version control”: he doesn't talk, he writes one line, but it works. Officials measured in the real Claude Code session that the average number of lines of code was 54% less, the number of tokens was 22% less, the cost was 20% less, the time was 27% less, and the security did not decline. About 1.24 MB.

15 万 Star、当天仍在涨的规则型技能包。它的设定很有趣——把那种「留着长马尾、戴椭圆眼镜、比版本控制还早入职」的资深工程师塞进你的 AI 里：他不说话，他只写一行，但它管用。官方在真实 Claude Code 会话里实测，代码行数平均少 54%，token 少 22%，花费少 20%，耗时少 27%，安全性没有下降。约 1.24MB。

I-have-adhd skill pack: make your coding agent answer first, explain lateri-have-adhd 技能包：让编程智能体把答案说在最前面

A small skill with around 52,000 stars that does exactly one thing: it makes the agent lead with the command, the file path and numbered steps instead of burying the answer in the third paragraph. The whole pack is about 226 KB and works with 20+ clients including Claude Code, Codex, Cursor, Gemini CLI and opencode, with official Simplified Chinese install docs inside.

约 5.2 万 Star 的小技能，只干一件事：让智能体先给命令、路径和编号步骤，再谈解释。砍掉「好问题」「希望有帮助」这类铺垫。整包约 226KB，兼容 Claude Code、Codex、Cursor、Gemini CLI、opencode 等 20 多个客户端，自带官方简体中文安装文档。

Anthropic's official open source financial intelligence skill library: ready-made methods for investment banking, stock research, private equity, and wealth managementAnthropic 官方开源金融智能体技能库：投行、股票研究、私募、财富管理的现成打法

The official Anthropic repository with approximately 38,000 Stars. It is not a model that makes decisions for you, but breaks down the most time-consuming repetitive work of financial analysts - building models, writing memos, and organizing research notes - into a skill package that can be directly executed by an intelligent agent, and the final judgment is still signed by a human.

约 3.8 万 Star 的 Anthropic 官方仓库。它不是替你拍板的模型，而是把金融分析师最耗时的重复劳动——搭模型、写备忘录、整理研究笔记——拆成智能体能直接执行的技能包，最终判断仍然由人签字。