daydreamer 🇺🇦

life is a blind box

[Rob Friedman @PitchingNinja]

Kai-Wei Teng's 3Ks in the 5th.

[Marc Andreessen 🇺🇸 @pmarca]

You two would like each other in person! @PalmerLuckey @CliffordAsness And yes both of your names are real, as surprising as that may be.

[atomic.chat @atomic_chat_hq]

OpenClaude from @gitlawb running through Atomic Chat's local model

MacBook Air M4 16GB RAM - blazing fast

No API key. No cloud. Fully open-source

[Mark Minervini @markminervini]

Today’s market strength was textbook. This is exactly what markets do during corrections when they get stretched to oversold levels. As I said just recently, "some of the biggest rallies occur during bear markets and corrections." Today was a perfect example.

Traders rushed in after headlines hit that Iran’s president signaled a willingness to end the conflict with the U.S. The Dow exploded higher by 1,125 points. But let’s not confuse cause and effect. The news may have been the trigger, but the market was already set up for a rally. It was oversold and primed. Now comes the part where discipline matters.

We ignore the first few days of a rally attempt. That’s potential noise. What matters is whether the market can follow through and whether leadership begins to emerge and proper setups develop.

Technically, this is a classic snapback: Indexes that broke below the 200-day are rallying back toward it, while Indexes that held the 200-day are bouncing off it. That’s typical countertrend behavior until proven otherwise.

Expect volatility to remain elevated. That’s not where low-risk money is made, but it's certainly where the risk is. Your job during corrections is simple: identify the stocks showing the best relative strength and the tightest price action. Those are your future leaders when the market finally turns.

On the macro side, nothing has been resolved. Higher crude prices are still a problem. Yesterday’s rally did nothing to materially bring down oil. The bigger issue is still in play and the jury still out. Oil at these levels feeds inflation, pressures growth, and gives the Fed a reason to stay on hold longer. Yields stay elevated in that environment.

To cut through all the noise, I look to the market itself, which has a much better track record of telling us the truth than the politicians, the analysts, the news, and the gurus.

The four steps of the bottoming process are:
1. Oversold – The difference between an ordinary pullback and an oversold condition starts with price, but it does not end there. Poor breadth and and a lack of volume confirmed follow through describe a one-sided market, and one not to trust.
2. Rally – Inevitably, the market bounces from its oversold condition. A high-quality rally is broad-based. A low-quality rally is defined by short covering and driven primarily by the stocks that have declined the most. Again, the character of the rally is important to distinguish. So far, we simply don't have enough data to make a confident determination, so patience is the watch word while we wait.
3. Retest – After the rally, there is almost always a retest. The popular averages approach, and in some cases breach, their oversold lows. The key to a successful retest is less selling pressure, such as fewer stocks below their moving averages, fewer stocks, sectors, and markets making new lows, less total volume, and less downside volume. If the retest fails, the process reverts and we generally start looking for divergences during lower lows. In the event of unexpected news, it is possible for the market to recover in a "V" fashion with no retest. In that case, we look at breadth confirmation and participation.
4. Breadth thrusts – In the final phase, not only do benchmark indices rally sharply with few pullbacks, but they do so with an extremely high percentage of stocks, sectors, and markets participating, or what technical analysts call breadth thrusts. In rare cases, the market has skipped step 3. With strong enough breadth, retests are not necessary. The Covid bottom is an example of a pretty powerful V-shaped recovery.

Bottom line:
This was an oversold rally, sparked by headlines—but not defined by them, and certainly not confirmation of a reliable bottom.

Now we watch:
--Quality of follow-through
--Emergence of leadership
--Market internals and model health

If the rally lacks quality, if economic pressure builds, or if leading stocks begin to deteriorate, then this remains what it likely is—a rally within a correction.

Stay objective. Let the market prove itself. If you are going to trade, do so incrementally.

https://t.co/JXzFFTmMtn

[麻婆豆腐 （エンジニア) @gyodon_se]

正直あまり言いたくないけど、
「Claude Code × Obsidian」で学習効率バグってる。

学びたいテーマ投げるだけで
・図解付きの座学
・初級〜上級のハンズオン
が一気に出てくる。

自分専用の技術書＋ハンズオン講座を作るSkillsって感じ

参考書とか講座買わなくて済むから
地味に節約できるのもデカい☺️

あ、今はAnsible勉強中です笑

[chiefofautism @chiefofautism]

someone built a Claude Code skill that CLONES ANY website from one command

its called ai-website-cloner-template, 4,800 stars in 16 days

you type /clone-website and point it at a url... it screenshots the site, extracts every design token, downloads all assets, then spawns PARALLEL agents to rebuild each section simultaneously

the output is a pixel-perfect Next.js clone, $0, open source

[Thariq @trq212]

To manage growing demand for Claude we're adjusting our 5 hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged.

During weekdays between 5am–11am PT / 1pm–7pm GMT, you'll move through your 5-hour session limits faster than before.

[Rez Karim @rezkhere]

always wanted something that could clip videos without me babysitting it...

turns out OpenClaw can do that now

drop a video link in Telegram (or any app you use)

get multiple viral shorts back in minutes, ready to post

captions in any language or style

Here's how 👇

[Simon Høiberg @SimonHoiberg]

Here's another quick way to make your OpenClaw agents forget less.

Internally, OpenClaw keeps a few files for itself:
- MEMORY .md (for long-term memory)
- /memory/[date] .md (for day-to-day memory)

And finally, the sessions files (jsonl).

How it's SUPPOSED to work:
The sessions files are used directly with ongoing chats. So OpenClaw will have context from the last few messages you did. The issue is, these sessions files get truncated as they grow, so it won't remember what you talked about a few days ago.

This is where the MEMORY .md and the /memory/[date] .md files become useful. OpenClaw will move important information to these files to store it as "long-term" memory.

The biggest issue I've found:
OpenClaw does a TERRIBLE job at deciding what to store in long-term memory. So it always ends up forgetting important things unless I explicitly tell it to store them in memory.

Here's how I fixed it:
I asked my agent to write a script that runs on an hourly CRON job. The script has one single function:

Go through the messy jsonl sessions files, clean them up (remove tools calls and other metadata), and write them to clean md files in a folder called "previous_conversations".

Then add a section to your AGENTS .md file instructing the agent to search through previous conversations when necessary.

⭐ Pro tip: If you have a RAG installed (highly recommended), have the CRON store the conversation chunks to the database as well, this will make it even faster and easier for the agent to search through.

After doing this, I ditched the standard /memory/[date] .md format altogether. Now I just have this + MEMORY .md for bigger, more highlighted information.

Try it!

[Shann³ @shannholmberg]

Karpathy's AutoResearch is changing how campaigns get optimized and most marketers haven´t heard of it yet.

Ole Lehmann tested it on landing page copy, 56% → 92% pass rate overnight.

here´s how it works for marketing / skills 🧵

[Conservatives @Conservatives]

Wow: Kemi dismantles Keir Starmer's surrender to China in less than 6 minutes.

[𝙺𝚒𝚖𝚋𝚊𝚕 𝙼𝚞𝚜𝚔 🤠 @kimbal]

The only way to apply sense to Trump’s trade war is to create massive uncertainty that will force the Fed to lower rates for when we need to refinance $6.5T of treasuries by June.

China knows this and by selling treasuries they create counter pressure for the Fed to keep rates high.

This game only needs to be played by China for 2 months until end of May when we need to refinance our debt.

Also remember that China is not a democracy and doesn’t have free speech. They can control messaging in their country while Trump cannot control it in America.

So.. Trump is playing chicken w the Fed. China knows this and is selling treasuries to try to make sure Trump loses the Fed-Chicken game and gets trounced by public opinion.

The Fed is run by patriots and will do what they can for America. But they are subject to market pressures like everyone else.

[Seattle Mariners @Mariners]

Look up the word “wholesome” and you’ll find this moment of Ichiro surprising one of his biggest fans 💙 #IchiroHOF

見た人たち全員を笑顔にする、イチローがファンにサプライズを届ける瞬間 💙

[daydreamer 🇺🇦 @nextplayer]

🇹🇼

[Jon Morosi @jonmorosi]

Yu-Min Lin on how the @premier12 gold transforms baseball in his country:

“It changes a lot. Now every Taiwan player — young guys, old guys — knows we can be champions too. We’re not just a normal country, a small country. Small countries can be world champions.” ⁦

[daydreamer 🇺🇦 @nextplayer]

😋

[吉野いくみ @ikumi_ysn]

仕事休んで北海道きた！ず〜〜〜〜〜〜っと夢だった「新千歳空港で食って飲むだけの日帰りツアー」はじめます！！！！

[daydreamer 🇺🇦 @nextplayer]

@zonghengjp 請問可以備份上傳 YT 嗎？ (會留下本連結)

[明和電機 @MaywaDenki]

AppleのCMを逆再生したら、めっちゃ明和電機のCMになった。

[ﾑﾗﾀ @murataart]

ポテトチップス揚げようと思ってモノタロウのカタログに挟んで水気きってた結果がこれ

[Warriors on NBCS @NBCSWarriors]

THIS IS NOT A DRILL 🚨

STEPH FULL-COURT TUNNEL SHOT 😱

[#AusOpen @AustralianOpen]

Genius at work 🧠

Su-Wei Hsieh smacking back a 200kmh serve like it's nothing 🤯

#AusOpen • #AO2024
@wwos • @espn • @eurosport • @wowowtennis