blablabla

[Serena @369Serena]

Serenity 火了以后，很多人第一反应是去抄作业。

但其实我们可以给 Codex 它装上这些投资类 Skills 和 AI 投研框架后

它可以变成一个股票研究助理：

帮你拆产业链、读财报、看情绪、查风险、做估值、写反方报告，甚至把策略拿去回测。

先收藏这 5 个 GitHub 项目!
别让 AI 直接替你买股票，先让它帮你把研究做扎实。

1️⃣ UZI-Skill⭐ 2.4K

中文投资者最值得先看的一个!

它把 A股、港股、美股分析做成了一套完整流程：深度分析、陷阱排查、龙虎榜、投资大佬评审团、22维数据、180条量化规则。

适合拿来做个股初筛、风险排查和中文市场研究。

GitHub：
https://t.co/1Dq2yXMLhq

2️⃣ TradingAgents⭐ 83.7K
GitHub 上最火的 AI 金融交易框架之一。
它把一个投研团队拆成多个 Agent：基本面分析、情绪分析、新闻分析、技术分析、研究员、交易员、风控团队。

如果你想看看“AI 投研团队”到底怎么协作，这个项目最有代表性。

GitHub：
https://t.co/N0VMCVroA9

3️⃣ Serenity Skill⭐ 731

最近 X 上讨论很多的投资 Skill。

它复用 Serenity 的核心研究路径：从热点出发，拆产业链，找供应链瓶颈，再回到股票和基金方向，检查公告、财报、客户、产能和风险。
适合研究 AI 半导体、CPO、机器人、算力、电力设备、创新药这类“产业链很长、信息很乱”的方向。

GitHub：
https://t.co/YGgiXR6aZr

4️⃣ Buffett Skills⭐ 512

这个 Skill 适合在你心动之前泼一盆冷水。

它会逼你检查：

这家公司是不是好生意？
有没有护城河？
现金流可靠吗？
管理层可信么？
价格有没有安全边际？

适合长期投资者做质量审查。

GitHub：
https://t.co/pDMtWc6LTB

5️⃣ QuantDinger ⭐ 7.4K

一个更完整的 AI 量化交易工作台。

它覆盖从 AI 研究、策略代码、回测、模拟盘，到实盘执行和监控的完整链路。

如果你不满足于“让 AI 分析股票”，想把想法变成策略并验证，QuantDinger 更适合继续研究。

GitHub：
https://t.co/cZqOdURO5u

别让 AI 替你买股票。

让 Codex 替你拆产业链、读财报、查证据、找风险、写反方报告、做回测。

这才是普通投资者最该抄走的用法。
#codex #skills #Serenity

[Nav Toor @heynavtoor]

🚨 Someone built an AI that reads candlestick charts the way GPT reads English.

Trained on 12 billion records from 45 exchanges. Outperforms every model by 93%. Live BTC demo. Free.

It's called Kronos.

The first open source foundation model built for financial markets. Not a general AI repurposed for finance. An AI that speaks the native language of candlestick patterns.

Every other model treats financial data like weather data. Kronos treats financial data like financial data.

Here's what it does:

→ Price forecasting. Feed it candlesticks. It predicts where price goes next.
→ Volatility prediction. Forecasts how volatile an asset will be before it happens.
→ Zero-shot. No fine-tuning. Works on any asset, any market, any timeframe.
→ 45 exchanges. Binance, NYSE, NASDAQ, LSE, and 41 more.
→ 4 model sizes. 4M params runs on a laptop. 499M for max accuracy.
→ Live demo running right now. BTC/USDT. 24-hour forecast. Updated hourly.

Here's the wildest part:

→ 93% more accurate than the leading time series model
→ 87% more accurate than the best non-pretrained baseline
→ All zero-shot. No fine-tuning. Out of the box.

Hedge funds spend millions on proprietary models. Bloomberg Terminal costs $24,000/year.

This runs on your laptop. Few lines of Python. Free.

Built at Tsinghua University. Accepted at AAAI 2026. Models on Hugging Face.

11.6K GitHub stars. 2.4K forks. MIT License.

100% Open Source.

[摸鱼巨匠🔨 @SunNeverSetsX]

这套在闲鱼卖 698 的 Claude Code 入门视频视频，是我见过最好的入门教程。

现在我不仅免费分享给你，而且还是中文版！

只要静下心看完这一套教程，你就学会用 AI 搭建工作流、实现自动化、处理复杂任务。

[Capital Flows @Globalflows]

These are THE TOP BOOKS I would give someone who wants to understand macro, interest rates, and FX:

🧵

Understanding the history of money

[Sac @Saccc_c]

外网疯传的斯坦福学生NotebookLM 48h 学习法 ，轻松帮你搞定所有大学课程的期末考试

他具体是怎么做的呢？

首先，他上传了关于这个课程的6本教科书、15篇顶级期刊论文，以及他能找到的所有相关课程讲稿。

然后，他提出了第一个问题：“这个领域的所有专家共同分享的5个核心心智模型是什么？”

这个问题的关键在于心智模型，这是教授们需要多年才能总结出的认知。

他之后提出了第二个问题：“现在告诉我，这个领域的专家在哪3个地方存在根本性分歧，每一方的论据是什么？”

通过这两个问题，他已经能明白整个这个学科的争论焦点、共识内容以及还未解决的难题。

接着，他提出最为关键的问题：“生成能真正区分深度理解者与死记硬背者的10个问题。”

然后，他会通过原始材料去回答这10个问题，并且每答错一个他会继续提问：“解释这为什么错了以及我遗失了什么？”

48后，他已经能和这个学科的老师进行深度讨论和辩论了。

AI时代，48h内你就可以成为这个领域的专家，学会如何系统性的提问才是你提升学习效率的关键。

[Himanshu sharma @pheonix_trader]

How to Identify good momentum burst and make millions by @PradeepBonde

Pradeep Bonde is a veteran trader who specializes in intermediate-term swing trading setups

Sharing Momentum Burst strategy framework & live example from Indian market👇

[关木 @ZeroZ_JQ]

Mac Mini 不够了，我要买一台大彩电了。

World Monitor这个开源项目，有点吓到我了。

把 170+ 新闻源、40+ 地缘数据层、军事/能源/金融/基础设施信号、AI 总结、3D globe 全塞进一个界面里。

冲突、航班、海运、卫星火点、金融市场、新闻流、甚至 AI 推演，都在同一张地图上实时流动。

更离谱的是：
可以 本地跑 LLM（Ollama / LM Studio）
支持 RAG headline memory
还能做 国家 instability index

[Brady Long @thisguyknowsai]

I DON’T UNDERSTAND WHY PEOPLE DON’T USE CLAUDE FOR STOCKS.

Most traders are looking at charts from 6 months ago.
CLAUDE analyzes real-time sentiment on X to predict future.

Here are 20 prompts to find the next 10X stock:

[blablabla @holynmoooo]

@DejaVu929292 “最好的权重品种，去年是银行，今年是保险”<----可以说一下rationale吗

[柴郡🔔｜Crypto+AI Plus @0xCheshire]

瑞·达利欧把自己公寓里的一个初创项目，做成了一个价值 1500 亿美元的商业帝国。

大多数人花一辈子都在试图理解市场。

而达利欧把 75 年的投资智慧，浓缩成了 42 分钟。

把这段视频收藏起来。

[Cyrille Jubert @Menthalo1]

China: Absolute master of the silver market
#China #Silver #LBMA #JPM #SLV
In this article, I argue that China may have implemented a strategy since January 2011 that has led to its current situation, where it holds all the cards.
China may hold, through intermediaries, almost all known silver reserves. Thus, China will be able to re-evaluate the price of silver relative to gold at will.

Is my hypothesis correct?
We will have the answer in a few weeks.

https://t.co/sl6N8xqEBl

[宝总的财富指南 @baozong_facai]

我不明白为什么更多人不用Google Gemini进行股票研究。

不是为了获取建议。
不是为了预测。

而是为了在冒险前理性思考。

这是我实际使用的10个详细提示 👇

[Pathfinder @Pathusa]

炒美股的朋友们注意了：电力、核电

川普宣布美国的数据中心将想办法自筹电力，而不是把成本转嫁到民用电，转嫁给老百姓。

接下来，估计各大厂和数据中心自建发电厂将成为主流，尤其利好天然气发电和核电厂商。

核电相关：OKLO、VST、CEG、TLN、DNN、NEE、GE

非核电 / 天然气 / 公用事业 / 可再生 / 现场发电：
NRG、D、ETR、AEP、XEL、BE、NFE、EQT、SWX、BEPC、FSLR

Grok选出的增长潜力最大的五只票

最有大涨潜力的Top 5（从高到低排序）：

OKLO（小型模块化反应堆SMR龙头）为什么最猛：Meta 1.2GW大单已落地，预付款+资金支持大幅降低风险；分析师平均目标价在100-110美元区间（最高175美元），当前市值仍有爆炸空间；SMR是AI长期零碳基载的核心叙事，如果首座反应堆2026-2027顺利上线，股价可能继续翻倍甚至更多。
风险：高波动，零营收阶段，监管延误可能回调。
潜力：短期到中期最有可能大涨（甚至10x+叙事）。

NRG（天然气独立电力生产商IPP）为什么强势：天然气是AI数据中心短期最可靠的“桥接燃料”，NRG有大量燃气资产+LS Power收购（2026 Q1完成，新增13GW产能）；数据中心合同扩展到1GW+；2026年利润预期超2025范围，分析师看好AI需求驱动。
近期已多次因收购/合同暴涨，2026年继续高增长。
潜力：2-5x空间，短期涨幅最稳健的高贝塔。

BE（Bloom Energy，燃料电池现场发电）为什么看好：解决电网瓶颈的“现场电力”神器，Google等客户+Brookfield 50亿美元合作；制造产能扩到2GW；2026年营收预期大增30%+，股价已从低点暴涨，但仍有上行（部分目标价80+美元）。
高估值但AI现场需求急迫，适合激进玩家。
潜力：继续大涨，但泡沫风险高。

ETR（Entergy，南方公用事业）为什么有戏：Meta在路易斯安那建10B美元AI数据中心，配套建1.5GW新燃气厂+41B美元2026-2029投资计划；工业/数据中心需求年复合增长13-14%；分析师视作AI电力受益首选。
稳健型，但真实项目落地后容易大涨。
潜力：2-4x，防御+进攻结合。

VST（Vistra，混合核+气）为什么还在榜：成熟玩家，Meta 2.6GW长约+核uprate；2026 EBITDA预期大增；分析师Strong Buy，目标价高（243美元+）。虽市值已大，但AI需求下仍有翻倍潜力。
比纯核稳，但涨幅可能不如OKLO/NRG激进。

[Annie 所长 @web3annie]

【2026开年爆炸股TOP32】最高涨幅 +107%！

1. $CRML 关键金属 Critical Metals Corp +107.89% ——稀有金属翻倍上涨，机构追捧
2. $UAMY 美国锑业 United States Antimony +63.24% ——工业稀有金属板块火爆
3. $SKYT 天水科技 SkyWater Technology +59.34% ——半导体制造成长股
4. $SNDK 闪迪 Sandisk +59.31% ——存储芯片需求爆发
5. $KTOS 克拉托斯防务 Kratos Defense +53.96% ——军工科技热度高
6. $BE 布鲁姆能源 Bloom Energy +51.12% ——清洁能源概念强势
7. $APLD 应用数字 Applied Digital +50.97% ——数字矿业与加密基础设施
8. $AVAV 爱维朗 AeroVironment +49.20% ——无人机与国防科技
9. $KRMN 卡尔曼 Karman +44.82% ——新能源与工业材料板块
10. $RCAT 红猫 Red Cat +44.73% ——航天科技成长股
11. $PDYN 帕拉迪纳铝 Palladyne Al +44.47% ——轻量化工业材料
12. $USAR 美国稀土 USA Rare Earth +44.46% ——稀土资源政策利好
13. $AEVA 艾娃 Aeva Technologies +42.11% ——激光雷达与自动驾驶
14. $NVTS 纳维塔斯半导体 Navitas Semiconductor +39.07% ——功率半导体应用广
15. $BKSY 黑天 BlackSky +38.59% ——卫星遥感与数据服务
16. $OKLO 奥科核能 Oklo +38.53% ——小型核能创新股
17. $EOSE 伊欧斯能源 Eos Energy +37.98% ——储能电池概念
18. $DPRO 德拉甘飞 Draganfly +36.71% ——无人机与智能监控
19. $ASPI ASP 同位素 ASP Isotopes +36.68% ——医疗与工业同位素
20. $SMR 新思能 NuScale Power +35.44% ——模块化核电技术
21. $NNE 纳米核能 Nano Nuclear Energy +33.04% ——小型核能成长股
22. $SERV 服务机器人 Serve Robotics +32.65% ——机器人自动化应用
23. $ASTS AST 天空移动 AST SpaceMobile +32.35% ——卫星通信潜力股
24. $ABCL AbCellera AbCellera +32.16% ——生物科技创新股
25. $FLY 火飞 Firefly Aerospace +31.73% ——商业航天发展加速
26. $ALMU Aeluma Aeluma +31.29% ——光伏新能源材料
27. $UUUU 能源燃料 Energy Fuels +30.22% ——铀与清洁能源板块
28. $ACMR ACM 研究 ACM Research +30.15% ——半导体设备创新
29. $ONDS Ondas Ondas +30.15% ——工业物联网与通信
30. $AXTI AXT Inc AXT Inc +29.30% ——半导体衬底与晶圆材料
31. $IREN IREN IREN +29.02% ——新能源成长股
32. $LTBR 光桥 Lightbridge +28.18% ——核能技术创新股

2026 年开局简直炸裂！🚀

[DAN KOE @thedankoe]

https://t.co/FGLYHaRhRq

[Wealth Code @WealthCode99978]

🌐⚡
“美国电力堆栈（U.S. Power Stack）”这份清单，已经不是投资清单，而是一张国家级能源与安全地图

这份列表真正有价值的地方，不在于“股票多”，
而在于它完整覆盖了美国未来 10–20 年能源安全的所有硬约束层。

如果把它拆开看，会发现逻辑非常清晰：
不是押某一种技术，而是为“电力 = 国家安全”搭建完整供应链闭环。

我按“功能层级”重新梳理一下这套结构。

⸻

🔋 第一层：核能是底座，不是选项

这里的关键词不是“发电”，而是可控、长期、不可替代。
•$CCJ｜西方反应堆体系的核心铀供应
•$NXE｜高品位铀矿，供给紧缺时的期权资产
•$UUUU｜美国本土采矿 + 燃料加工，去地缘风险
•$LEU｜先进反应堆不可缺少的浓缩能力
•$UROY｜对铀产量的特许权杠杆
•$LTBR｜延长现有反应堆寿命的燃料技术

这一层解决的不是“便宜”，而是持续运行与政治安全性。

⸻

⚛️ 第二层：核系统 + 国防绑定

核不只是电力，而是国防与战略威慑的一部分。
•$BWXT｜海军核动力与国防核系统
•$MIR｜辐射探测、核安全
•$ASPI｜国防 / 医疗关键同位素
•$OKLO｜面向数据中心与国防的微型反应堆
•$NNE｜下一代核部署路径

这里的共同点是：
订单来自政府，节奏慢，但粘性极强。

⸻

🔌 第三层：电网与电力韧性（真正的瓶颈）

再多电源，没有电网都是空谈。
•Tesla ($TSLA)｜公用事业级储能骨干
•$EOSE｜长时储能，解决“核能 + 可再生”的时间错配
•$BLDP｜燃料电池，用于备用电源与关键设施
•$SEI｜电力 + 储能基础设施整合
•$AMPX｜航空 / 国防高密度电池
•$MVST｜军用与车队级电池

这一层的核心不是技术炫技，而是电力不断供。

⸻

🔋 第四层：锂——所有电气化的物理入口

不管你信不信电动车，只要信“电”，就绕不开锂。
•$ALB｜全球锂供给锚
•$LAC｜西半球锂资源战略布局
•$SGML｜低成本盐湖
•$SLI｜直接锂提取（DLE）技术

这里解决的是：
谁能在本土体系里掌控电池原料。

⸻

🧱 第五层：战略金属 = 隐形护城河

这是最容易被忽视，但最难替代的一层。
•$MP｜稀土与永磁，军工与电机核心
•$USAR｜美国本土稀土
•$UAMY｜锑，弹药与高端合金
•$FCX / $FM｜铜，电网与电气化血管
•$CRML｜多金属战略敞口
•$IVN｜工业级基础金属
•$METC｜钢铁用冶金煤
•$AMRK｜实物金属分销

没有这些，“能源转型”只存在于 PPT。

⸻

🧠 真正的结论

这不是在押“新能源”，
也不是在押“核能”。

这是在押一个判断：

👉 未来十年，美国最重要的资产不是应用，而是“电力可控权”。

而这份 Power Stack 清单，本质上就是：
•能源安全
•国防独立
•AI 与算力扩张
•工业回流

四条主线的交集。

问题只剩一个：
当资本真正意识到“电力=国家安全”时，
你觉得最先被重估的，是哪一层？

#USPowerStack #EnergySecurity #NuclearEnergy #Uranium #GridResilience #EnergyInfrastructure #StrategicMaterials #RareEarths #Lithium #DefenseStocks #AIInfrastructure #NationalSecurity

[SpacBobby @SpacBobby]

$ASTS - How 6G satellite could bridge coverage gaps on smartphone

Source - Pakistan Press International

By 2030, according to a comprehensive analysis, mobile networks could integrate satellite capabilities so seamlessly that users wouldn't notice whether their signal comes from ground infrastructure or spacecraft racing overhead at orbital velocities, according to StudyFinds' publication.

International standards bodies are advancing specifications through successive technical releases, though deployment typically lags years behind specification completion. Companies like SpaceX, AST SpaceMobile, and Lynk Global have proven direct satellite-to-phone connectivity works in field tests, and smartphone makers are beginning to incorporate satellite modems into devices. The goal: extend coverage to areas cell towers can't economically reach and reduce 'no service' dead zones in remote regions, at sea, and during disasters.

StudyFinds said the performance targets are ambitious. While today's 5G networks deliver peak speeds of 10 gigabits per second, the upcoming 6G standard (officially called IMT-2030) targets 1 terabit per second. Low-orbit satellites currently offer latency around 20-50 milliseconds, far better than older high-altitude satellites at 250 milliseconds, though still not quite as fast as ground-based networks.

Coverage would extend to areas cell towers can't economically serve like remote villages, cargo ships crossing oceans, planes flying over polar regions, and disaster zones where infrastructure has been destroyed.

According to StudyFinds, the fundamental challenge with current mobile networks is economics. Cell towers work well in cities and suburbs where high customer density justifies infrastructure costs. That math falls apart in rural areas where expensive equipment serves far fewer people spread across vast distances. Large populations worldwide remain without reliable internet access, concentrated in these underserved regions and developing nations.

Satellite systems have existed for decades, but they operated as separate networks requiring specialized equipment. Think satellite phones costing thousands of dollars, used mainly by ships, military personnel, and expedition teams. The shift involves integrating satellite capabilities directly into mobile networks that already power standard smartphones, allowing devices to switch between terrestrial and satellite connections.

The International Telecommunication Union (ITU) has set 2030 as the target for the 6G networks. The specifications require not just faster speeds and lower delays, but seamless integration between ground-based cell towers and satellites orbiting at various altitudes. Standards are advancing through 3GPP releases, though years typically pass between specification freeze and mass deployment.

The publication noted that three types of satellites will handle different jobs. High-altitude geostationary satellites (positioned about 22,000 miles up) stay fixed above one spot on Earth and can cover enormous areas (up to 2,000 miles wide) making them ideal for network management and control. Medium-altitude satellites operate between roughly 1,200 and 12,000 miles up, providing regional coverage and services like navigation. Low-orbit satellites, flying at just 300 to 1,200 miles altitude, circle Earth every 90 minutes and deliver response times of 20-50 milliseconds.

SpaceX has already launched over 5,000 Starlink satellites into low orbit. Amazon's Project Kuiper and OneWeb are deploying competing networks.

According to the study, published in the journal Engineering, these 'mega-constellations' represent a shift in satellite economics: instead of launching a few expensive satellites that must work perfectly for 15 years, companies now mass-produce simpler spacecraft that can be replaced more frequently. When one fails, dozens of others provide backup coverage.

The trade-off is coverage area. Each low-orbit satellite's footprint spans just 60 to 600 miles across and remains visible for only minutes as the satellite races overhead. Achieving global coverage requires thousands of satellites working in choreography, with users' devices transferring between satellites as one disappears over the horizon and another rises.

On the technical building blocks, the report noted that standards groups have spent the past five years advancing specifications, with details still being worked out for eventual deployment around 2030 and beyond. The work involves solving problems unique to moving satellites.

High-speed satellite motion creates Doppler effects that shift radio frequencies. Modern systems track satellite positions and adjust frequencies in real-time to compensate, similar to how an ambulance siren changes pitch as it passes. Getting this wrong means garbled signals and dropped connections.

Laser links between satellites represent another major advancement. Early satellite systems routed everything through ground stations, creating bottlenecks and gaps over oceans. Satellites now communicate directly with each other using laser beams that achieve 400 gigabits per second, more than 1,000 times faster than older radio links. Data can hop from satellite to satellite, often crossing continents with minimal ground contact until reaching its destination.

Managing transitions between satellites presents ongoing challenges. Devices must smoothly transfer connections as satellites move, with systems calculating optimal targets based on trajectory, signal strength, and capacity. Recent specifications include algorithms designed to predict handoff timing and adapt to usage patterns to minimize disruptions.

The applications for 6G satellite technology extend beyond phone calls in remote areas. Autonomous vehicles could operate with consistent connectivity rather than only in areas with 5G coverage. Emergency response could benefit when natural disasters destroy ground infrastructure, satellite-integrated networks could help restore basic communications more quickly by routing traffic through orbiting spacecraft while crews work on repairs.

Cargo ships could offer crews reliable broadband for communication while transmitting real-time monitoring data from onboard sensors. Airlines could provide more consistent WiFi service regardless of flight path. For developing nations, the technology offers potential to deploy hybrid systems where satellites beam internet to strategically placed ground stations serving surrounding communities, potentially reducing the need for extensive traditional infrastructure.

According to the report, the 2030 timeline for 6G faces uncertainties. Radio spectrum allocation remains contentious as satellite and terrestrial networks must share limited frequencies while avoiding interference with each other and aviation systems. International regulators continue working on coordination mechanisms.

Orbital debris poses growing concerns. SpaceX alone has regulatory approval for over 40,000 satellites, more than 10 times the total number of satellites launched in human history. Each becomes a potential collision hazard. Companies are designing satellites to burn up in the atmosphere after their useful life, but managing tens of thousands of orbital objects requires unprecedented coordination.

Power consumption presents technical challenges. Modern satellites that process data in orbit rather than simply reflecting signals demand substantial electricity beyond what solar panels easily provide. Add in power-hungry laser systems and beam-steering capabilities and satellites push available power limits.

Artificial intelligence (AI) will likely play increasing roles in managing these networks. Machine learning could help predict optimal routing through constantly shifting satellite positions, anticipate handoff needs, and allocate spectrum based on real-time demand; potentially allowing networks to adapt to actual usage patterns rather than following rigid rules.

StudyFinds said some smartphone manufacturers have begun incorporating basic satellite capabilities, enabling emergency text messaging when cellular coverage disappears. These early implementations are limited but demonstrate that mainstream phones can incorporate satellite modems without major size or cost increases.

Whether the industry meets the 2030 target depends on resolving spectrum disputes, managing orbital traffic safely, and finalizing technical standards while allowing time for deployment after specification freeze. Standards groups are advancing through releases, but years typically pass between specification completion and mass market availability. If successful, however, the technology could substantially reduce the distinction between 'having service' and 'being on Earth.'

[blablabla @holynmoooo]

@taoray 我也是在23年底,24年初写了一大篇分析推了黄金给我老板。。结果投资没被采纳，老板也没觉得我有add credit🥲

[f #7inMelbourne @fvaeey]

😭😭😭

[The Tennis Letter @TheTennisLetter]

Darren Cahill says Jannik Sinner watches Carlos Alcaraz matches more than any other player

“Jannik watches more Carlos matches than he does anybody else. Because he’s fascinated with the improvements coming in his game. And he’s pushing us as coaches to make sure that he’s improving also as a tennis player. The rivalry is real. It’s there. Hopefully it’s going to be there and real for the next 10 or 12 years.”

(via Wimbledon Press)