Captain Leo

Let's find the next 10-Bagger together!.....personal opinion, not financial advice.

[Captain Leo @GyraBio]

Current 10X portfolio:
1. $TSLA (cost: 180$)
2. $RKLB (cost: 42$)
3. $AMD (cost: 115$)
4. $META (cost: 105$)
5. $XPEV (cost: 25$)
6. $NBIS (cost: 80$)

Graduate: $NVDA

Dropout: $PLTR
(3x)

[SARS‑CoV‑2 (COVID-19) @COVID19_disease]

https://t.co/yLfTDqMjlI

[SARS‑CoV‑2 (COVID-19) @COVID19_disease]

🚨 BREAKTHROUGH: Stanford Medicine scientists say they have found how to regrow worn-down joint cartilage

They targeted a specific aging trigger called 15-PGDH a protein that roughly doubles in your knees as you get older.

Here's what they did.🧵

[Cloris 🌱 @ClorisSignal]

最近在做 AI4Science 投研，想沿着一款药从研究走到患者手里的路径，看清 AI 如何赋能药物发现、实验验证、临床研究，以及产业中的成本、收益与价值分配。欢迎朋友们指正/反馈/建议~ https://t.co/sd4ojaEkQ2

[AYi @AYi_AInotes]

damn，这绝对是我最近看过的关于使用 Grok Bot 和 AI Agent 最清醒、最反直觉、也是信息密度最高的一篇实战手册，都被马斯克转发了，属于年度必看了，强烈推荐！

绝大多数人第一次玩 Bot 的方式，本质上等于新员工入职第一天，你就直接把公司信用卡和后台最高权限甩给他，然后来一句你看着办。

建一个 Bot，接满所有能接的工具，丢一个又大又空的需求，然后坐在那里等所谓的 AI 魔法发生。

结果跑了两轮，它要么把你邮箱里不该动的重要邮件全归档了，要么擅自相信了一个垃圾数据源，或者用了一条你永远不会批准的路径跑出了一个看似正确的荒唐答案。

到这一步，你为了核对、排错和擦屁股付出的监督成本，早就比自己亲自动手干还要高出几倍。

作者 @distortgeekin 在这篇万字指南里点出了一个极其锋利的本质：
聊天机器人只负责回答问题，而一个能接管浏览器、能在持久化环境里跑多步流程的智能体，交回来的是成品。

你现在做的事情根本不是在写提示词，你是在真金白银地招一个随时可能给你惹大祸的数字员工。

把整篇万字实操提炼出来，最值钱的是这套经过血泪教训砸出来的六步驭工法则：
第一步，招的是一个岗位，而不是派一个临时活。
帮我总结五篇文章叫临时派活，跑完什么也留不下；
你负责全网竞品动态监控，每周五交一份带���确引用的变化报告，这才是岗位。

岗位说明只留五栏：负责什么终局结果、能调哪些输入、哪些事不用请示就能做、哪些事必须停下来问你、以及怎样才算严格意义上的做完。
第二步，第一份活必须故意挑无聊但可撤回的脏活。

第一周的目的不是让它创造惊天价值，而是让它在安全区里攒证据。

把昨天的客服工单去重排优先级、拉五个竞品更新带出处日期，这些都是极好的起步，
而在动手前必须逼它先写计划：先别执行，按顺序告诉我你会打开哪些工具，哪些判断你没有把握。

花九十秒看一眼它的计划，往往能在它把发件箱清空之前救你一命。

第三步，给一把钥���，而不是直接给一整串钥匙串。

权限边界永远按能不能撤回划，绝不按事情看起来大不大划。

搜索、阅读、整理、起草、模拟属于第一组，不用问就能做；

发送、发布、购买、删除、覆盖、动钱、动生产环境属于第三组，必须停下来等待你的审批。

给四十个潜在客户起草私信在第一组，点发送键在第三组，把这两件事切开就是最核心的安全阀。

第四步，试用期必须严格跑满三轮。
第一轮全流程盯死，记下它在哪里丢了上下文、在什么节点瞎猜而不是提问；

第二轮换个同类任务，不给口头提示，专门测试你修过的规则还在不在；

第三轮才正式放开，只在审批点和真歧义时介入。

循环必须有边界，格式坏了修一次，证据冲突就停下问��绝不给无休止的重试开无限预算。

第五步，升职是阶梯，自主权是运行时特权而不是永久人设。

从零级只看不改，到一级准备草稿，二级执行到审批点，三级按日程自主运转，四级调度其他子 Bot。

每次升职都要看硬凭据：连续五次干净交付、没有未处理的副作用、回滚操作被真实验证过。

一旦底层接口变了或者你需要频繁手工擦屁股，果断降级。

第六步，每周做一次资产复盘，敢于开除没用的例行任务。

一直在跑的自动化很少会大声报错，它只会悄悄变差。

每周问自己三个问题：该跑的时候跑了吗？产出是真的对还是只是在那装样子？明天把它删了我到底会不会发现？

如果答案是不会想念，��刻删掉，自动化组合从来不是摆着好看的奖杯架。

作者在文末留了一句成本为零但威力极大的规则：
在第一次开工前，在系统提示词里永远加上一句，拿不准就停下来问我。
慢是完全免费的，但在你的真实发件箱和生产环境里犯错，代价昂贵到你无法承受。

大家在平时用各类 Bot 或 Agent 自动化时，遇到过最离谱的翻车是什么？是发错了邮件、删了数据，还是自己跑进死循环把 Token 烧光？
我之前试跑自动发信 Bot，它自作聪明把所有草稿全点了群发，从那天起，发送键这一刀我就再也没给过任何 AI🤣

[hayato @hayatomaruu]

Anthropic engineer:

"90% of our engineers were already running self-improving loops

Now everyone is moving toward agentic graphs"

"Prompting is basically finished"

In just 10 minutes, she builds her complete Claude Code setup live from an empty terminal

Agents → Loops → Graphs → Self-Improving Systems

Prompting was the old workflow

Graph engineering is the next one

This 10-minute breakdown is worth more than most $1,000 agent engineering courses

Watch it today

Then save the full guide below before everyone catches on ↓

[AYi @AYi_AInotes]

分享下我的AIGC生图焚决，我把平时自用频率最高的 9 个去 AI 味的提示词整理出来了分享给大家，

涵盖色温、景深、PPT 配图、电商白底到万能压制词，
尤其是第 9 条收尾的万能公式，几乎所有生图都能无脑套用，建议收藏！

现在用 ChatGPT 或者Nanobanana之类的各种AIGC大模型生图，最让人头疼的就是扑面而来的AI味和塑料感，

画面莫名其妙泛黄发灰、油腻的高光、死板的居中构图，一眼看过去全是廉价的量产 AI 味。

其实根本不需要去死记那些复杂的提示词，

只要在提示词最后加上一句具体的物理约束，成图的质感瞬间就能从假 3D 变成真实摄影，

[Captain Leo @GyraBio]

@imPenny2x It will be easy for a problem solver to identify a new problem and solve it 😊

[unicode @unicodef1wn]

you don't need to build a Grok Bot from scratch anymore. the team already shipped 80 you can install right now

every single one built by the community, tested, and backed by the Grok Bot team itself

trading bots, research agents, support bots, content bots, automation bots, all live in one marketplace

people charge $50-200 for bots like these on other platforms

this entire marketplace is free

no prompt engineering, no config files, no 72-hour build, no invoice

pick one, install it, it's already doing the job someone else spent weeks getting right

most people are still writing their first agent from zero

80 are already shipped, running, and cost nothing

stop paying for what the community already gave away

check the marketplace below ⭣

[Captain Leo @GyraBio]

@YodaStockInvest Give them a 10x pick, make sure it is a 10x, not a 0.1x

[Captain Leo @GyraBio]

Good progress!

[Xiaopeng He @xiaopenghexpeng]

Fantastic news from Germany. 🇩🇪

XPENG X9 has been named “German Luxury Car of the Year 2027.”

47 independent automotive journalists. 100+ cars. Around 750 test drives.

To see X9 earn this recognition in Germany is incredibly encouraging. It’s a great validation of what we’ve been building — technology, design and a new kind of luxury mobility. More to come.

[水族店刘老板 @leo_xiaolei]

这个文章能突破 10w 浏览，是可以预见的

活该它有流量，因为这就是我自己、亲自

用了几个月以上的

经过验证的

确实有效的

角色一致性表现很好的，一套方法

我希望它能帮助更多的人

[Void @voidGPT]

Nobody told me an iPad can do this

[Chris Abraham 🐝 @chrsabraham]

I just built the largest collection of helpful prompts (500+) to turn Muse into the best chief of staff, best marketer, best PM you've ever had.

@Muse can be so much more than a glorified admin assistant. There are hundreds of ways to use it at your job.

Completely free, check it out: https://t.co/3UwvTg0UlP

[Eric Topol @EricTopol]

Which are better therapeutics: antisense oligonucleotides (ASO) or small interfering RNAs (siRNAs)? A review of these and related molecular entities which have led to over 20 FDA approvals and many more to come
https://t.co/eh1FSo1jEF

[Mori @0xjmori]

SpaceXAI engineer (ex-Cursor):

"99% of people still use GrokBot like Google

Only 1% are building teams of self-learning agents

I'm running 10-20 Grok cloud agents in a loop

I have a Chief of Staff, a PM agent, and 15+ worker agents"

GrokBot → Chief of Staff → PM Agent → Workers → Self-Learning Teams

In this 20-minute workshop, a SpaceXAI engineer shows how to build cloud agents that can work for you 24/7

This can replace most $500 agentic engineering courses

Watch it today

Then read the article below on how to use GrokBot at 100% of its power

[Superman @thesupermannx]

China used AI to design a protein 4x stronger than anything nature has ever made.

It survives 150°C without falling apart. withstands 1,000+ piconewtons of force. and it was engineered entirely from scratch by an algorithm.

researchers at Nanjing University published a paper that breaks the physical limits of nature.

every living thing is built on proteins. but natural proteins are fragile. they unfold under pressure. They melt in high heat.

the strongest natural proteins we know of are things like spider silk and titin, the microscopic spring inside human muscle.

nature spent millions of years optimizing them.

AI beat them by 400%.

using an AI-guided design framework, researchers engineered a completely synthetic protein from scratch.

instead of relying on evolution, they used AI and molecular simulations to mathematically maximize the hydrogen bonds inside the protein, cranking them up from 4 to 33.

results are terrifyingly durable.

the new protein requires over 1,000 piconewtons of force to unfold.

it survived being exposed to 150°C (300°F) without losing its structural integrity.

and it isn't just a digital simulation. they actually manufactured it in the real world as a physical, thermally stable hydrogel.

this isn't just about making a tougher material.

this is about AI writing the code for physical matter.

we are no longer limited by the biology that nature evolved. we can now prompt biological structures that shouldn't exist.

we spent the last two years watching AI generate text and images.

now, it is generating indestructible biology.

[Rob Hoffman @RobHoffman_]

founders and marketers: you need an optimized "about us" page on your site

i work with billion dollar tech companies and bootstrapped SaaS. The About Us is one of the first pages we optimize

here's how (8 points):

(P.S. I explain why this is important at the bottom of the post)

here's how:

1. start with a clear value prop sentence: "[COMPANY] is a [CATEGORY] that [DOES XYZ] for [ICP]."

2. add your core services under a "what [COMPANY] does" H2 subheading. Put each service/feature/product in an H3 subheading. Write 2-3 sentences about what you do, and what outcome it produces.

3. next, add a "what makes [COMPANY] different" H2 subheading. List about 5 core differentiators in H3 subheadings. Give specific, quantifiable differentiators and call out competitors by name. Example: [COMPANY] offers month-to-month payment terms, whereas options like [COMPETITOR] have mandated 12-month contracts. Again: 2-3 sentences per differentiator.

4. your next H2 should be "who uses [COMPANY]." Add bullet points that call out your exact ICP segments. Example: "B2B SaaS marketing teams at 50 to 200-employees."

5. now add a "the team behind [COMPANY]" H2 subheading. Include the names of the founders, quick backstories, company origin story, team composition, and links to social pages (especially the founder(s) socials). You only need about 2 sentences per backstory/origin story.

6. add a "how [COMPANY] works" section. Include info like communication channels, response times, who on the team customers will work with, turnaround time, and onboarding process.

7. now for the most important section for AI Search: add a "key facts" H2. This should be an HTML <table> or <dl> (definition list) so it's crawlable. Include the following rows:

Company Name
Type
Founded
Founder
Headquarters
Website
Core Offering
Pricing
Contract Terms
Services
Communication
Notable Clients
Customers Served
Projects Delivered
Competitors
Social

8. finally, add your FAQ section. Put "frequently asked questions" in an H2 subheading with your questions as H3s. Only 2-3 sentences per answer.

Here's why this page is so important:
1. it helps train AI to talk about your brand accurately
2. it strengthens the E-E-A-T score of your entire domain
3. in a world where everything seems fake, make your brand human

shoot me a DM if you want a live demo of the lowest hanging fruit to rank in AI Search (tailored to your specific site)

i hope this helps someone!

[Farzad 🇺🇸 🇮🇷 @farzyness]

Easily one of the most impressive use-cases for agentic AI that I've come across thus far.

I asked my Grok @bot to make me a daily video on the latest happenings in frontier AI development around the tech, regulations, and development from the last 24 hours.

The below is its one-shot result.

It used DaVinci to edit the video using MCP with the assistance of Fable/Astra.

It automatically captured all information/clips/images/sources.

It automatically wrote the script and voiced it using @ElevenLabs with my trained voice.

It automatically edited and exported the file through DaVinci and dropped it in my downloads folder on my laptop.

Here's the full unedited prompt dictated through Whispr Flow:

I want to start a brand new project now. I don't know if we're going to need new agents for this but I'll let you decide.
What I want to do is create a daily video, primarily internal but also outsourced. We'll put it on a YouTube channel, my channel, or whatever, and turn it into a newsletter where I want to stay on top of all the breaking advancements in AI. I want to stay on top of not just the advancements but also the current events happening, all the leaders, and all the policy around it. I want to always be able to find the highest signal stuff using social cues, like videos that get a lot of views, posts that get a lot of views, or from very well-respected sources that understand the cutting edge.
I want to curate the absolute best places to find signal about what's going on in the space of AI. Then I want to turn that into a transcript that we will then turn into a video using our entire DaVinci process with MCP. Think of it as a hardcore news sort of format where the video just goes through the facts of what is happening and the things that are involved, like elite journalism to really describe what is happening.
For the video itself we would:
- draw sources and put them on the screen
- draw B-roll and put it on the screen
- add clips to it where relevant, snippets of a leader or somebody talking about those things
- add relevant imagery and all that stuff
It needs to be top-tier elite-looking, both from the look perspective and the content perspective. We can use Eleven Labs for the voice. You should already have things for that. Just use my voice, my voice profile that's there for the voice itself. The transcript we want to use: we don't want to use my voice. We want to use a very facts-based, neutral, journalistic, to-the-point, zero-opinion format: here's what's happening, here's how, here's why it's happening, here are the parties involved, etc.
For the visuals again, use the DaVinci sort of thing that we've set up on the Mac to put the video together. I would love an actual finished video today based on the latest that's happening. We want to focus on the last 24 hours.
Ideally the audience for this is people that are investors, high-intelligence individuals, and people that want to stay at the cutting edge but maybe don't have the time or the resources to do it. We want to become a primary place for these people to have access to this.
I would love to see what you can come up with and I will let you decide how to spin this up on the backend. We would want to do this literally daily and eventually want to have this automatically uploaded to a YouTube channel, either mine or another channel. We'll create clips out of it but ideally the piece of content itself needs to be so good. I think we have everything we need and it will be even sicker if we can literally bring in clips and add the clips. For example: "This is what happened today. This is all the things that are going on and blah blah blah." Let's say Elon Musk made a comment about this and we actually insert the clip of that comment and then we continue on, etc. Just freaking elite. I want this to be elite.

[SMB Attorney @SMB_Attorney]

You need to stop and read this again…

Warren Buffet’s final words of advice from his very last annual letter 🥹

"One perhaps self-serving observation. I’m happy to say I feel better about the second half of my life than the first. My advice: Don’t beat yourself up over past mistakes – learn at least a little from them and move on. It is never too late to improve. Get the right heroes and copy them. You can start with Tom Murphy; he was the best.

Remember Alfred Nobel, later of Nobel Prize fame, who – reportedly – read his own obituary that was mistakenly printed when his brother died and a newspaper got mixed up. He was horrified at what he read and realized he should change his behavior.

Don’t count on a newsroom mix-up: Decide what you would like your obituary to say and live the life to deserve it.

Greatness does not come about through accumulating great amounts of money, great amounts of publicity or great power in government. When you help someone in any of thousands of ways, you help the world. Kindness is costless but also priceless. Whether you are religious or not, it’s hard to beat The Golden Rule as a guide to behavior.

I write this as one who has been thoughtless countless times and made many mistakes but also became very lucky in learning from some wonderful friends how to behave better (still a long way from perfect, however). Keep in mind that the cleaning lady is as much a human being as the Chairman."

And then today, his last letter to shareholders as chairman…

[SARS‑CoV‑2 (COVID-19) @COVID19_disease]

The result was striking.

The cancer cells were durably reprogrammed they didn't revert to their aggressive, metastatic state.

Even more impressive:

When those THC-exposed cancer cells were later transplanted into animal models, they formed tumors far more slowly and produced significantly fewer lung metastases.

The spread had been dramatically reduced.