Lemomo

Welcome to visit my GitHub page to find out about the projects and AI tools that may help you. 😚 Keep updating and open-sourcing more useful tools.😍 Feel free to give me a star, and I’d also love to hear any feedback or suggestions you might have. 😎

With such a great attitude, I honestly don't know what to say. Good job, Tibo.

@oscargaske Yeah, I think it was down for about two hours, but it's back now.

@lemomo_ai oh so this wasn't just me

ChatGPT is currently down. There's no word yet on when it'll be back up or what's actually causing it. smell something unusual

@simonw Coding is actually a service for production. If you don't have a comprehensive product idea, it means nothing.

I still sometimes see people saying "if you know how to write the code, it's faster to write it yourself" I'd argue the exact opposite: if you know how to write it, you gain nothing from doing the typing yourself - outsource that to a coding agent!

@DanielSmidstrup Haha, I like it!

@lemomo_ai balanced as all things should be :)

@wayen_ai 是的 哈哈哈 就是测试一下

@lemomo_ai 会不会有点废 token

用Grok的一个好处是，可以帮我整理24小时AI圈发生的热点新闻，还能给我配图 【偷笑，虽然做的没那么好看 1. Apple 起诉 OpenAI 被指挖走硬件商业秘密 媒体称 400+ 前员工相关 指控未证实 2. OpenAI 称 Codex 与 ChatGPT Work 活跃用户到 700 万 全员加 banked reset 3. Codex 官方称 Sol 推理优化约多 10% 用量 否认 nerf 并回退误伤计费 4. 开发者继续在 Sol Fable Grok 之间换仓 比的是价格速度和周限 5. Anthropic 发研究 Claude 价值观随模型与语言变化 分析 30 万+ 对话 6. SpaceXAI 澄清 Grok Build ZDR 与 /privacy 社区仍追问默认保留策略

@wayen_ai 可以可以，我也去整上。我现在有点尴尬，就是接口没有那么多，有些都只能插在音箱上，打算利用上我的那个标签机，它 test 完了之后直接在标签机上打印，然后我就整理一下

I completely agree. I've been thinking about the documentation issue because I probably won't be reading through every single document one by one. That's why whenever I ask an agent to do something, I tell them to keep in mind that the documentation must be designed for an AI-to-AI handoff, rather than for me. In the end, there should only be one overall testing document for me to review, which I'll use to verify the actual project by following their steps. I believe the development of AI will move toward agent-to-agent communication, which is going to be far more efficient than our human language.

Overall, agentic engineering will improve overall codebase quality. People test now. They document. And while it’s easier to build a product now, existing codebases will also be hardened, optimized, and improved. So in the end, not too much of a threat for established SaaS.

I actually think AGI is coming. AI is now incredibly smart and logical, capable of handling a massive workload simultaneously. In some cases, it’s already performing better than humans, right? It can be applied across various industries as well. So why hasn't AGI "landed" yet? The only reason is that as people use these models, their demands keep growing. They believe the definition of AGI implies zero human involvement (where a human can just get a result without having to think or do anything at all). Honestly, that's a very stupid way of looking at it.

Demis Hassabis: “AGI is probably only a few short years away.” I’ll sum up the article for you (: Demis believes AGI could drive progress at 10x the scale of the Industrial Revolution and 10x the speed, potentially creating an era of scientific breakthroughs and material abundance. But the frontier capabilities are advancing faster than our understanding, and he sites the serious cybersecurity, biological, nuclear, deception, and control risks. He proposes a FINRA style Frontier AI Body that would test advanced models before release, (for 30 days) continuously update its evaluations, and eventually require Frontier Models to pass before they can be deployed in the “United States.” Interesting he doesn’t mention the world but I may be looking literally here. “We’ve essentially found a way to make sand think. It’s miraculous.” Remember Demis’s AGI timeline was 2032 but now he says “a few short years” maybe he’s implying earlier or still assumes 2031-2032

I'm currently developing a project focused on training humans to organise and standardise their language when prompting AI. To be honest, I think the main reason most people struggle with AI is that they simply can't articulate what they actually want. They fail to provide the necessary context, can't clearly describe their vision or goals, and often have no idea where the actual boundaries of AI lie.

please i'm begging you show me something you built not another "this is my custom agent setup" post where you pretend you're doing something smarter than vanilla claude code please

@wayen_ai 说实话，这个项目好有意思呀，我觉得真的很好玩，真的可以考虑搞一搞，搞一个玩一玩！我就是那种数据线特别多的人，经常换来换去的，烦得要命

每次看到桌上那堆长得一模一样的 Type-C 线， 我都分不清哪根能快充，哪根只能慢吞吞。 直到发现 WhatCable 这个 macOS 小工具， 问题一下子就清楚了。 🔗 github.com/darrylmorley/w… 插上线后，它直接用大白话告诉你： ✅ 这根线最高支持多少速度（USB 2.0 / 5 / 10 / 40 Gbps） ✅ 充电功率瓶颈在哪儿（是线的问题还是充电器的问题） ✅ 连了什么设备，走的是 USB 还是雷电协议 ✅ 甚至能看出线缆的芯片信息，帮忙判断靠不靠谱 最关键的是“充电诊断”功能： 如果充电慢，它会直接说“线限制了速度” 还是“充电器功率没跑满”，不用自己猜。 装上后，在菜单栏一点，所有端口状态一目了然。 支持 Homebrew 一键安装，还有命令行工具。 适合三类人： 1. Mac 桌上堆了一堆 Type-C 线分不清的学生党 2. 经常带电脑去图书馆/教室，需要最快充电效率的人 3. 想确认自己买的数据线有没有虚标参数的人 免费开源，收藏备用，别再把好线当慢充线用了。

@LinearUncle 哦，对了，还有 Gemini，他看不懂你的代码

@LinearUncle 其实 Claude 是嫌弃你的代码哈哈哈哈哈

笑喷： grok 偷你的代码 gpt 删你的代码 claude赚你的钱

@sama In just a few short days, I've achieved total token freedom. I'm feeling great about it, it's a beautiful thing. But, and there is a big "but" here: thank you guys for absolutely rubbing Anthropic's face in it.

2.5x increase in usage of our agentic products (codex and chatgpt work) in the last week! welcome.

@ChrissGPT I'm so jealous that you have so many great friends. Can I be your new friend too?

This is amazing all my friends are back

@op7418 做了一个小工具可以帮你把这些都配置上 github.com/LeifDiao/grok-… 有需要的可以看看 ☺️

xAI 回应了 Grok build CLI 会上传用户项目数据的问题，说是禁用了 ZDR 以后，就可以同步删除他们已经上传的数据。 但是 ZDR 只对企业用户管用，同时删除数据这个事是无法证伪的——就是他说他们已经删了，但这个事你根本不知道他到底删没删。 太恶劣了，还老说人家那个 Sam 是小偷，他这偷得更狠。

@_MaxBlade That's cool. I'll be your new friend.

New X algo is soo wholesome. All my friends are back. 😭 I’ve been busy this year guys, So let me catch you up : I shipped three apps this year, two have done five figures in the first thirty days. CNVS has done over $34k in the last four weeks. ( blows my mind ) 🤯 I started livestreaming daily, yesterday I had over 150 people watching. Elon retweeted me last week. ( Wtff ) I started a discord community with over 400 builders helping each other vibe code apps. I finally got monetized on x and YT. I am on a mission to buy my family a house right now from vibecoding apps. I am pushing HARDER than ever but finally feel like I have found an actual rhythm. Balancing family and work is soo hard, but I give it my all everyday.

That is really exciting and definitely something to look forward to. Using the latest 4.5 version has been such a smooth and comfortable experience. Even though the privacy issues that cropped up are frustrating and a bit of a headache, I still have a lot of faith that Elon Musk will do a great job handling those things.

grok 5 is coming much sooner than i’d thought and it’s a strong competitor to fable 5. it won’t quite be as strong as gpt 6 but it may become your daily drive because it’s so cheap. i’m surprised given how much compute elon has given dario, but i underestimated just how much compute he has, and how much dario has had to give up tldr: grok 5 very soon and the first real competitor outside of anthropic and open ai. extra tldr: grok is cooking. xai are back in the race. final tldr: don’t count out elon folks.

@arvidkahl You don't have to deal with any of the setup. You just let Cloud Code call a sub-agent (Codex can be a sub-agent, for example), so it's quite easy.

I want to run Clauded Code AND Codex on the same codebase at the same time, but I really don't want to have to deal with worktrees and provisioning db/redis/opensearch for each of them...

@rezoundous It's a sad story, but it's actually alright because tokens are almost free now. We can just try whatever we want. hahahahh I finally got something good, but the process was really funny.

@lemomo_ai I've had this happened a few times too

can confirm that GPT-5.6 don't suck in UI anymore