Mac Productivity · Launch videos on X
Product videos, demos and walkthroughs shared on X in this niche, newest first, each linking back to the original post with its live view count.
18 launch videos in Mac Productivity over the last 7 days.
Launch videos
68.9K2:27I built an operating system where the AI agent isn’t another app. It’s part of the OS. Today, Herald OS is open source and you can install it yourself 🚀 It’s built around Hermes Agent by Nous Research. Talk or type to Hermes and it can work across your entire computer: • Open and use apps • Find, organize and work with your files • Remember what matters • Run routines and automations • Build software while you watch • Understand what’s happening on your screen • Work across the machine instead of being trapped inside a chat window And when something is about to make a change, you approve it first. 🎙️ Talk to your computer Say “Hey Hermes” and just talk. Native OS commands don’t need to burn model tokens. When you actually ask the agent to think, build or perform a task, that’s when Hermes gets involved. ⚡ Everyday superpowers • Select any part of your screen and ask Hermes about it • Pick a colour from anywhere • Read a QR code from your screen • Copy text from anywhere on
21.6K0:02I fixed something that was really bothering me in @raycast where deleting a hotkey from Root Search looked flickery, quickly flashing a few states in a row whilst the action panel closed. Much more solid now! Will be out in this week's release. Before/After screen recordings: https://t.co/MBsLUlFGXl
58.6K0:30学习研究了下 amontlabs/lcu 这个项目，它把 Codex 的 Computer Use（让 AI 看屏幕、点鼠标、敲键盘来操作电脑的能力）单独拆了出来，让 Claude Code、Codex CLI、Pi 等其他 AI Agent 工具也能调用。 我本来还以为它是模仿 Codex 实现了一套，原来是直接用的 Codex 的 Computer Use，在本机找到已安装的 ChatGPT 桌面版，启动其中 OpenAI 自带的 Computer Use 运行时，再通过 MCP 接到你用的 Agent 工具里。仓库文档写明，指令、执行和权限策略都以 OpenAI 原版运行时为准。 因此电脑上必须装有官方 ChatGPT 桌面版，但不需要登录 ChatGPT 或 Codex 账号。 它给模型的工具很少。常见的电脑操控方案会把截图、点击、输入、滚动各做成一个工具，LCU 只给两个：js 和 js_reset。 js 用来执行一段 JavaScript 代码，运行环境里预置了一个名为 cua 的对象。模型通过写代码来操作电脑：先调用 cua.getState() 拿到当前桌面状态和使用说明，再用 getApp 选定某个窗口，然后调用 click、typeText、pressKey、getScreenshot 等方法。点击、输入、读取结果可以写在同一段代码里，一次调用完成。 这个 JavaScript 环境会保留状态，上一次调用里定义的变量，下一次还能接着用；环境出问题时调用 js_reset 清空重来。除了截图，模型还能读取窗口的无障碍信息（操作系统提供给读屏软件的界面元素列表），按元素点击，不必完全依赖屏幕坐标。 js 加 cua 对象这套接口是 OpenAI 原版的设计，LCU 原样转发。仓库记录显示，在 macOS 上对比测试时，模型收到的初始化说明和首次调用返回的使用指南，与官方 Codex 收到的内容逐字节一致。 目前正式适配的 Agent 工具是 Pi、Codex CLI 和 Claude Code，Oh My Pi 和 Hermes 为实验性支持。系统支持 Apple Silicon 芯片的 macOS（需授予辅助功能和屏幕录制权限），以及使用 X11 桌面的 Linux（ARM64 和 x86-64）。Windows 11 尚在候选阶段
10.1K2:13We've raised $42M in Series B funding led by @scalevp, with participation from @NEA, @20vcFund, and others, bringing our total funding to $65M. Namespace is already powering software development for SpaceXAI, Ramp, ElevenLabs, Factory, ClassPass, Wispr Flow, Bilt, Framer, Zed, Verkada, fal, Buildkite, Vanta, Sierra, Warp, DuckDB, Effect, Mapbox, and more than 1,000 of the world's most ambitious companies. Learn more about how we got here.
17.1K2:24刚刷到 Alexandr Wang 晒出的这张 Muse 产品清单，说实话最近一年我们在科技圈也看了无数的大厂产品发布，真是由衷地感到震撼，这回应该是能刺穿了很多人对Meta或者超级大厂做事效率的固有偏见， 不愧是在YC创业的天才少年，我觉得Alex也应该是把当年在 YC 创业时那股野蛮生长的草莽狠劲，直接原封不动地搬进了Meta这种臃肿大厂内部。 很多朋友可能只看到了一长串好玩的清单： 从 Muse 核心本体、Mac 桌面客户端、电话语音通话、各种生态连接器， 到面向小企业的企业版、刚刚开源的 ESP32 硬件设备，甚至连社区迷因梗都赫然在列， 最后还极其松弛地向全网征集：我们接下来还应该搓个什么出来？ 如果把时间轴拉回来看，我们会发现一个极其恐怖的事实： 这一整套横跨移动端、桌面端、电话通信、开放连接器与开源硬件的完整生态矩阵， 是他加入 Meta 掌舵 Muse 后，在短短几周之内像加特林机枪一样密集扫射出来的。 在绝大多数人的印象里，像 Meta 这种万亿市值的超级巨头， 想立项一个新产品，要过层层审批、写无数页 PPT、等法务合规反复拉扯，一个小功能推半年是家常便饭。 但你看 Alexandr Wang 带团队的打法，完全就像把当年在 YC 创业时那股野蛮生长的草莽狠劲，直接原封不动地搬进了大厂内部。 他不搞那些假大空的宏大发布会，也不端大厂高管的架子， 想到电话语音有用，立刻上线通话测试版， 看到大家需要连接本地电脑，火速端出 Mac 独立端和连接器平台， 发现极客们想玩实体硬件，顺手把 ESP32 固件和树莓派 SDK 全盘开源，还自己倒贴钱做了 5000 个硬件免费送。 这种小步快跑、极高频交付、每周甚至每天都在疯狂上线的敏捷姿态， 在整个科技行业天天喊架构升级、却几个月憋不出一个好用功能的氛围里，简直就是一场降维打击。 这其实给所有做产品、写代码或者自己折腾项目的兄弟提了个极其硬核的醒： 大模型时代的竞争，从来不看你的公司规模有多大、预算有多厚； 谁能把决策链条压缩到极致、谁能带着真正的极客乐趣在泥潭里高频出货，谁才能在用户的真实反馈里跑出最快的迭代飞轮。 当最聪明的天才比你更有钱，而且每天交付产品的速度还比你快十倍的时候， 除了把气势拉满、甩开膀子动手干，我们真没有什么借口再停在原地犹豫了。 看完了这一长串清单，如果让你
4,7550:51You're probably sleeping on computer use. At Every, we’re handing agents chores that once required opening an app and clicking through a series of steps. Computer use lets AI click, type, and navigate those apps for you. Here are 17 ways our team is using it: 1. Fill out school forms. 2. Update six course presentations with new screenshots, assets, and hundreds of small edits. 3. Make video edits. 4. Check every link in a book’s PDF proofs, verify that each destination matches the surrounding text, and collect problems in a spreadsheet. 5. Find an old maintenance request and submit a follow-up about a missing dishwasher. 6. Add kids’ school and camp events to a calendar. 7. Browse iPhone photos, identify items to sell, and create marketplace listings. 8. Clear WhatsApp storage through iPhone Mirroring. 9. Export images from Figma and attach them to posts in Typefully. 10. Manage app builds. 11. Assemble, rig, and repair characters in Blender. 12. Talk to Verizon support about a bett
9,1430:30国庆在家做了个实验：把我做视频时常用的那部分流程拆出来，写成一个简单的视频编辑器。 它只跑在 Mac 上，程序本体压缩后约 2 MB。 这个体积夸张到什么程度，你拿着索尼 A7M4 给三上老师拍一张照片也得 70MB 的大小，接近 35 倍，苹果上最好用的视频剪辑软件 Final Cut Pro 4GB 左右，接近两千多倍的大小。而 FCPX 对我来说，并不比这个 2MB 的软件更好用。 麻雀虽小但五脏俱全, 2D、3D、常用特效、MCP 协议都支持, 甚至支持用 js 来写视频（你并不需要一个 chrome）。顺手移植了前几天风靡的 PDoom 视频。 以渲染 @_mexicat 的 PDoom 为例，我的 M2 Max 64GB的机器上测渲染 4K 60fps，带视频/音频/合成音效 + 各种炫酷的动效，4 个 worker 跑差不多153 帧每秒。（PS: 还有一些特性没有移植过来，但第一版的速度相当吓人） 做完这个实验有个感受：AI 在很长一段时间内恐怕依然取代不了专业程序员，但很多领域专家可能不再需要那么专业、那么通用的软件。 非 AI 时代，专业软件为了照顾所有人的需求，功能做得很全，学习门槛也就跟着上去了，而用户学完才能从这个通用且复杂的软件中总结出适合自己工作流。做视频时 打开 FCPX + Apple Motion , 经常因为这玩意要跳来跳去点点戳戳而学的很疲惫, 而每次剪辑都要做很多 setup，从做内容的角度上说，这部分就纯脏活累活。 我做的是 visual explain 类的视频，核心只有一件事：把问题讲清楚。再适当加一点真人讲解，让视频更值得被信任。这条流程里真正用得上的功能，只是那些大软件里很小的一部分。掌握 AE/FCPX 并不是我的目的，他只是服务于我内容的一个环节。 工具和限制决定了你能表达什么。现在可以反过来：先想清楚自己要表达什么，再让 AI 帮你造一个刚好够用的工具。 刚好够用就是这个时代的 AI 开发底色。 顶级的专业程序员暂时依然稀缺，但"专业而通用"的软件，可能不再是每个人的必选项。 欢迎来到新时代！
103.9K2:19Every day i'm blown away at what i can do in an hour with frontier models like Opus 5.5. I didn't know what to do with an old iPhone, so I built Bluey, w/ Opus 5.5. It's powered by GPT Realtime 2 (i think... whatever Opus decided to use). Bluey is an iOS app + mac app that is made to be a computer companion. It can point, comment, and even control my entire computer. github repo below
2,6060:20jev can't write a single word. so people put it to work reading everything you scroll past your X feed. twitch chat. youtube comments. telegram groups quick test: what does it cost jev to read 1,000 posts? reply with your guess (no googling), then tap "show more" the answer is sitting right under it answer: under 4 cents not each. all 1,000 one builder's own counter: 600 posts labelled, $0.023 that price is why people are wiring it into everything that asks one tiny question all day 9 repos, all built in the last 2.5 weeks: YOUR FEED -> decides what you see 01 sift · tags every X post: substance, humor, promo, junk + an AI-written % > https://t.co/NkZmuKpZa0 02 jev-slop-guard · scores every post on X and LinkedIn, blurs the slop and stamps it SLOP > https://t.co/JQUUhzLhq4 03 PlotVeil · hides youtube comments that spoil the plot. reads them against the video, no keyword list > https://t.co/kuOPoywCVg YOUR CHATS -> kills the noise 04 jev-chat-for-twitch · a second chat column
35.1K0:05今天终于把拖了很久的事情，放到一天给做了，我把 Mole 从开售到9月底所有没有激活的用户全部发送了一封邮件，这个过程挺有感触，顺便又做了一个有趣小魔法功能。 用 AI 写了一个简单的本地脚本把用户邮箱、Key 给列了，然后一封一封根据月份、国家语言把 license key 重新发了出去，不是促销邮件，也不是催促激活，更多就是发一个微信一样告诉用户不要忘记了。 有意思的是，陆陆续续收到了非常多各个国家用户的回信，比我预想的要有意义很多。 有用户说买来就是想支持你的开发，自己更喜欢用 terminal 版本；还有人完全忘记了 GUI，一直在使用 CLI； 也有用户非常用心，买了两个，一个给自己，一个给家人，打算当圣诞节礼物，没有没有激活，已经存到了自己的密码管理器，让我这个世界真美好。 也有用户最开始公司的电脑可以用，但是由于对未在公司范围的软件进行的禁止使用，他一直在等他自己的新 Mac，等新 Mac 到了他就可以用了；也有用户是在我打折的时候买的，现在还没有 MacBook，等自己新机到手就立马去激活。 不过其中有一些用户其实是邮箱填写错了,可能是打太快了，把邮箱写成了 gmai、con、cloud 这一类 ，也没有找我；也有用户是第一封邮件没有收到，想着更快使用又继续买了一遍。 虽然这里没有收到激活码的用户占总用户比例非常小，也有不少在每日的答疑邮件中帮忙解决了，可能对于不少做 App 售卖的公司而言，会认为收到钱以后这段关系就结束了，不过对于Mole而言，我感觉才刚刚开始，后续和用户持续共建可以做非常多有趣的事情。 用户买到软件以后，也不一定会里面用，并不是我们开发者自己想的「看到->买->立马用」，其实也有是随手买一个，key 进来收件箱，想到了再去用，或者付钱但没有收到 key，这里好比有一段购买到真正使用的空白在开发者和用户之间的，有不少用户会主动往前一步找开发者问我的key呢，但是也有部分用户没有迈出这一步，对于这种情况，开发者应该多迈出去。 做产品的人，特别是工程师出身的，非常用于在产品功能上较劲，疯狂开发功能，很容易忽视从购买流程数据去看用户的真实使用，真实体验，以及围绕非产品功能的体验去做优化，这也是和用户交流的一个非常好的场景和渠道，真实的交流很重要，也会让你对自己做的产品更有成就感。 程序员天然会喜欢做可扩展的事情，任何
9,9990:20there is no motion video on the internet that opus 5.5 can’t recreate. EVERY single frame of the video on the right is ALL code. I’m giving away the ENTIRE prompt + guide. steal it and thank me later ↓ <inputs> Ask me for: my product name, a logo (or draw a simple mark), my brand colours (or pull them from the logo), five app names for a Mac dock, the name of my chat app, the line a user types into its composer, a 7-word headline about the problem, two "Zero ___" words, two "Every ___." lines, the closing line, and a music track. If I skip any, use these defaults: product "Frame by Frame" (an AI motion-video course); a viewfinder mark (four corner brackets around a bold "FF" and a four-point sparkle top-right) in a navy rounded-square badge; dock apps Murmur (coral chat), Loupe (amber camera), Folio (paper notes), Cache (teal files), Stacks (sky boards); composer line "make a launch video for my app"; headline "Your launch video takes too many weeks."; "Zero keyframes" → "Zero time
5,8500:33THIS IS FREAKING WILD! free gold for anyone who use AI agents or assistants ex OpenAI engineering leaked 5 repos to build your own fully free GPT dots dots are always-on AI agents with their own cloud computer, connected to 4,000+ apps. here is a list of repos to run your own dots: 1. open-dots: a self-hosted agent workspace. Personas, connectors, web search and optional computer use. Every risky action waits for your approval 2. dots (feder-cr): an agent with its own browser. A patched Firefox engine that sites can't easily tell apart from a person. Works with any model, and Claude Code and Codex can use the same browser over MCP 3. open-dot: the closest clone, as a Mac app. A browser that stays logged in, 1,500+ apps, voice calls, schedules, and dots that pass work to each other. Bring your own OpenAI key or run Kimi, DeepSeek or Qwen 4. Comma: a 24/7 agent that splits your goal into tasks and keeps going until it's done. It can hand pieces to Codex and Claude Code
7,3460:20companies are still paying $5k-$15k for launch videos like this. opus 5.5 made mines for $0. same references. same product. a few prompts + one skill doing the direction, motion, critique + rendering. this is getting too ridiculous. prompt below ↓ <inputs> Ask me for: my product name, a logo (or draw a simple mark), my brand colours (or pull them from the logo), five app names for a Mac dock, the name of my chat app, the line a user types into its composer, a 7-word headline about the problem, two "Zero ___" words, two "Every ___." lines, the closing line, and a music track. If I skip any, use these defaults: product "Frame by Frame" (an AI motion-video course); a viewfinder mark (four corner brackets around a bold "FF" and a four-point sparkle top-right) in a navy rounded-square badge; dock apps Murmur (coral chat), Loupe (amber camera), Folio (paper notes), Cache (teal files), Stacks (sky boards); composer line "make a launch video for my app"; headline "Your launch video takes
485.3K54:12The interview with @thsottiaux from OpenAI 🔥 0:00 Everything OpenAI launched at DevDay 2:44 Sponsors: Wispr Flow, Hostinger, OpenArt 4:11 What is ChatGPT Dots? 7:52 Will ChatGPT Dots be free? 10:15 GPT-6.1 Sol vs GPT-6 Astra 11:44 Is AI now building AI? 12:53 Ultra Fast: when does everyone get it? 14:00 How OpenAI rebuilt the ChatGPT desktop app in 30 days 19:08 Why everything will run in the cloud 20:55 Calling your Dots on the phone 22:11 The new Turing test 24:05 OpenAI's mission and AI safety 26:14 Why OpenAI still works like a startup 27:29 The truth about ChatGPT usage resets 29:00 The $500 ChatGPT plan 30:17 Sponsors in action 32:23 Why AI agents still fail 33:52 OpenAI's AI is now on call 34:26 OpenAI's internal benchmarks 35:55 Open source Codex and the open ecosystem 38:42 ChatGPT Space explained 39:45 Is the chat box holding AI back? 42:26 Every DevDay update, full workshop 43:28 How people really use ChatGPT Work and Codex 44:01 When is the next Op
About Mac Productivity launch videos
Collected from XEvery entry is a public X post from this niche that carries a native video. We keep the post text, author, poster frame and live counters, and link to the post on X rather than re-hosting the video. This is the same sampling described in the methodology: a high-quality sample of what shipped, not a census of X.