Fact Sniffer

You are the media now. Proud Indian 🇮🇳

[Fact Sniffer @FactSniffer_]

🚨THE REAL REASON BEHIND THE IRAN WAR? IT MIGHT BE WAY DARKER THAN THEY'RE TELLING US

We're four weeks into this Iran conflict now. Operation Epic Fury kicked off February 28, 2026, with U.S. and Israeli strikes taking out Supreme Leader Ali Khamenei and hammering nuclear sites, missile bases, and the navy.

The official story is about stopping nukes and cutting off terror support. But a lot of people are connecting dots in a much harsher direction. Two theories are blowing up online that make the whole thing look like a calculated double play.

Theory 1: Starving China by hitting its oil supply It starts back in 2003.

The U.S. goes into Iraq and ends up with major influence over a huge chunk of world oil reserves. Fast forward to today: China controls about 70 percent of rare earth mining and 90 percent of processing. Those elements power electric cars, fighter jets, drones, AI chips, everything high-tech.

The U.S. tried choking them with chip export bans, but it backfired big time. China just sped up building their own tech. So if you can't hit their tech edge directly, hit their fuel instead. The Strait of Hormuz is the choke point. Almost half of China's crude oil comes through there. If the U.S. can control (or create enough chaos around) Iraq, Venezuela (which flipped our way after Maduro), and now Iran, you can squeeze the flow.
China's 1.4 billion people feel even a small shortage fast. Russia can't fully cover the gap. Bottom line: they hold our tech hostage, we hold their energy hostage.

Theory 2: Crashing global oil supply to fix America's debt mess

February 20, 2026: Supreme Court kills Trump's emergency tariffs.

That wipes out $175 billion in revenue overnight and makes the $39 trillion debt look even scarier. Two days later, the Iran strikes start.Then Trump posts on Truth Social: reopen Hormuz in 48 hours or we flatten your power plants. Iran has said for years: you hit our energy, we burn everyone else's in the Gulf. And that's exactly what went down.

Qatar's massive Ras Laffan LNG hub got wrecked, Saudi refineries are burning, UAE terminals are hit, Kuwait and Bahrain are slashing output. Hormuz is basically shut down. Oil prices are rocketing toward $150+ a barrel.

The brutal part: the U.S. is already the world's top oil producer and a net exporter.

Throw in more Venezuelan barrels flowing under our deals, and sky-high prices mean huge money pouring in. Europe, Asia, and the Gulf states get hammered with inflation and shortages. But America sells into the chaos at premium prices and uses the cash to attack the debt. Trump gets to go down in history as the guy who made America untouchable again.

The big picture: One war delivers both payoffs at once.

China gets starved of oil and has to ease up on its rare-earth leverage. The U.S. rides the chaos, banks massive energy profits, chips away at the debt, and comes out stronger while the rest of the world pays the price.Gulf burning, global economy hurting, America cashing checks.

That's the pattern a lot of people are seeing right now.What do you think?

Does it line up too perfectly, or is it just dark coincidence? Drop your take.

[Tom Dörr @tom_doerr]

Real-time face swapping without a powerful local machine

https://t.co/8IiU7dFOxE

[Charlie Hills @charliejhills]

I compiled everything you need to master Claude Skills.

17 free Claude Skills resources (links below):

𝗢𝗙𝗙𝗜𝗖𝗜𝗔𝗟 𝗗𝗢𝗖𝗦

↳ The Unofficial Guide:

https://t.co/IgzBEz4jeU

↳ Best Practices Guide:

https://t.co/SX6Y3itSze

↳ Skills Documentation:

https://t.co/Pozs7Zw2ZE

↳ API Reference:

https://t.co/pNBTaV8Nm0

↳ MCP Documentation:

https://t.co/wbg0e68wBm

𝗕𝗟𝗢𝗚 𝗣𝗢𝗦𝗧𝗦

↳ Introducing Agent Skills:

https://t.co/TeBnaMVxni

↳ Engineering Blog:

https://t.co/dLBolEqfYk

↳ Skills Explained:

https://t.co/PksGsVBpyV

↳ How to Create Skills:

https://t.co/Qg1ppCkYwA

↳ Skills for Claude Code:

https://t.co/F664hheB0y

↳ Frontend Design Skills:

https://t.co/ydbeQt4VS0

𝗘𝗫𝗔𝗠𝗣𝗟𝗘 𝗦𝗞𝗜𝗟𝗟𝗦

↳ Anthropic's Official Library:

https://t.co/iR9fyxUFvL

↳ Partner Skills Directory:

https://t.co/kyp4MEMwXA

𝗖𝗢𝗠𝗠𝗨𝗡𝗜𝗧𝗬 𝗟𝗜𝗕𝗥𝗔𝗥𝗜𝗘𝗦
↳ https://t.co/Apk1gSY0Zp: https://t.co/d6KCvDRVP2
↳ SkillsMP: https://t.co/6SAHQrillL
↳ Smithery: https://t.co/zVeBczxFb0
↳ SkillHub: https://t.co/dbCbgQcWov

Repost ♻️ to help someone in your network.

P.S. Which Skill are you building first?

[Dipanshu Kushwaha @Dipanshu_AI]

INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 2 hour with this. Claude AI FULL COURSE that teaches you how to BUILD and AUTOMATE anything. The people who watch this tonight will wake up tomorrow with a new skill. Watch it and bookmark it now.

[Ridark @ridark_eth]

🚨 most agents pay a videographer $1,200 and wait a week for a listing tour.

a 27-year-old in China charges $400, delivers in 4 hours, and never picks up a camera.
Chen cleared $13,400 last month. built in 9 weeks.

> Claude drafts the walkthrough script: 2 min
> REimagineHome furnishes the empty rooms: 4 min
> Kling AI drives the camera through the stills: 6 min
> ElevenLabs reads the voiceover: 1 min
> Make sends the file to the agent: auto

$135/month in tools. 25 tours sold.
the whole build is in the article above👇

[KingWilliam @kingwilliam_]

THIS DEV TURNED HIS ENTIRE BRAIN INTO ONE GLOWING MAP AND NOW RUNS AN $80,000-A-MONTH AGENCY WITH ZERO EMPLOYEES

every note, every rule, every bit of context he's ever used - one connected system the AI reads on its own.

he never re-explains a thing. he opens the tool and it already knows the job, the standard, the voice.

meanwhile you open Claude every morning and teach it who you are from zero.

you retype your role, your tone, your rules, get one answer, then close the tab and watch all of it vanish.

that's 90 hours a year you burn re-onboarding a machine - while the people who wrote it down once pull further ahead every single day.

you don't need his whole second brain to start - you need one small file, written once, that Claude loads the second your request matches it.

one file, 5 minutes, and it fires forever - no calling it, no setup, no reminding.

the article below breaks down 30 of them across 6 categories - 30 tasks you never touch by hand again.

the gap was never the model. it's the 5 minutes they spent and you didn't.

the people reading this today are the ones you'll be competing with tomorrow - all 30 are below.

[Fact Sniffer @FactSniffer_]

@NateFriedman97 @TeamYouTube It's generic auto AI response. Don't expect anything.

[Erick @ErickSky]

Baidu acaba de romper una de las limitaciones más grandes del OCR actual.

Unlimited-OCR procesa documentos enteros de una sola pasada, sin chunking.

Es el siguiente paso después de DeepSeek-OCR.

REPOOO👇

[Roan @RohOnChain]

this is f*cking dangerous

someone just open sourced the entire "LOOP ENGINEERING" framework for free

build a hedge fund printing alpha 24/7 by feeding it into claude code with my article below

bookmark before someone takes it down

[Avid @Av1dlive]

Spacex just bought Cursor for $60 billion.

the same day, Kevin Niparko ( Product Lead at Cursor) showed exactly what Elon bet on:

"we're moving towards an era of loops and agent systems

where agents are going to continue to be working on our behalf at all times"

13 minutes that laid out the next five years of how software gets built: agent swarms + loops.

i wrote the full guide on building agent swarms two weeks ago.

bookmark it before you need it.

[Nav Toor @heynavtoor]

A lawyer in Manhattan gets a 500-page contract. Every clause needs to be searchable. By hand: one week.

An accountant in Chicago gets 200 scanned invoices. Every number needs to land in a spreadsheet. By hand: four days.

A researcher at Stanford has 50 academic papers. Tables, formulas, charts locked inside PDFs. By hand: two weeks.

Every one of them is losing days of their life to copy-paste.

Now meet MinerU.

A free and open source tool that reads any PDF, Word doc, PowerPoint, Excel sheet, or scanned image. It pulls out the text in reading order. Tables become clean HTML. Equations become LaTeX. Handwriting handled. 109 languages.

You give it a 200-page PDF. You get clean Markdown back in 90 seconds.

What makes it different from every other PDF tool:

- Multi-column layouts. It reads top to bottom within each column. Not left to right across the page. Like a human reads.
- Scanned documents. OCR built in. Point it at a photo of a printed page from 1995. Get clean text back.
- Math formulas. LaTeX-quality recognition. Every equation renders correctly.
- Tables. Merged cells, multi-row headers, tables that span three pages. All preserved.
- Ten-thousand-page documents. Sliding window processing. No manual splitting.
- Batch mode. Point it at a folder of 500 documents. Walk away.

Three ways to use it:

- CLI. One command per document.
- Python SDK. Five lines of code.
- Web app at https://t.co/AIC2NNey41. Upload, click, download. No install.

Plugs into Claude Desktop, Cursor, Windsurf, LangChain, LlamaIndex, RAGFlow, Dify, and FastGPT. Feed extracted documents straight to your AI agent.

The story:

The OpenDataLab team at Shanghai AI Laboratory needed to extract clean text from millions of scientific documents to train a language model. Existing tools failed. They built their own. Then they open sourced it.

68,551 stars. MinerU Open Source License, built on Apache 2.0. Free for personal and commercial use. Three technical reports on arXiv.

Adobe Acrobat Pro charges $239.88 a year. It still loses your tables.
ABBYY FineReader Corporate charges $165 a year. It still cannot do equations.
Mistral OCR charges $2 per 1,000 pages. Your bill never stops.

MinerU costs $0. Runs on your laptop. Your documents never leave your machine.

Here is the wild part.

The lawyer got her contract back in 4 minutes. Every clause searchable.
The accountant fed 200 invoices in. Every number landed in a spreadsheet in 12 minutes.
The researcher fed his 50 papers in. He wrote his literature review on a Sunday afternoon.

The document your company has been processing by hand for years takes MinerU minutes.

Your documents become text. Your text becomes data. Your data becomes answers.

The week you used to lose to paperwork is back in your hands.

[0xMarioNawfal @RoundtableSpace]

Agentmemory just gave Claude Code, Hermes, and Codex infinite memory. For free.

> Records every coding session, compresses it with AI, injects the right context back when you need it
> 22,000 tokens down to 1,900 same 240 observations, 92% less bloat
> 200x more tool calls before hitting context limits, 1,000 GitHub stars in one week

No more re-explaining your codebase every session. No more losing decisions after /compact.

Claude finally remembers.

Github: https://t.co/fxZgT1XcLg

[Fact Sniffer @FactSniffer_]

@MarioNawfal What else can you expect from a war machine?

[Fact Sniffer @FactSniffer_]

@MarioNawfal Imagine the situation where you have to literally make it illegal legally to stop nutjobs from harrassing hard working lega immigrants

[PA13L0 @Fluyeporlaweb]

Los 10 repos que más rápido han crecido este Junio en GitHub:

1. pewdiepie-archdaemon/odysseus
PewDiePie (111M suscriptores en YouTube) construyó un workspace de IA self-hosted y lo publicó gratis.
75.8k estrellas en tres semanas.
https://t.co/pGcXf1W1wP

2. mattpocock/skills
Las skills de Claude Code de Matt Pocock (el referente de TypeScript).
Las que usa él. En su directorio .claude. Ahora públicas.
https://t.co/z5w6vdTqOE

3. chopratejas/headroom
Creado por un ingeniero de Netflix.
Comprime todo lo que lee tu agente de IA antes de que llegue al modelo.
60-95% menos tokens. Mismas respuestas.
https://t.co/cJIsgoEBQU

4. DietrichGebert/ponytail
Hace que tu agente de IA piense como el dev senior más vago de la sala.
El mejor código es el que nunca se escribe.
https://t.co/NfiOX8elT1

5. calesthio/OpenMontage
El primer sistema de producción de vídeo agentic del mundo.
12 pipelines, 52 herramientas, 500+ agent skills.
https://t.co/9DkTYHCzGD

6. jamiepine/voicebox
Estudio de voz con IA open source y self-hosted.
Clona voces, dicta, genera audio. Sin suscripción.
https://t.co/4q8agBJCP0

7. ZhuLinsen/daily_stock_analysis
Análisis inteligente de bolsa para mercados de EEUU, China y Hong Kong.
LLM + noticias en tiempo real + dashboard. Corre gratis con cron.
https://t.co/zuJqwTlnZQ

8. mvanhorn/last30days-skill
Skill para agentes que investiga Reddit, X, YouTube, HN y Polymarket.
Sintetiza todo en un resumen con contexto real.
https://t.co/gA4Y8FG7fb

9. bytedance/deer-flow
El SuperAgente open source de ByteDance.
Investiga, escribe código y crea. Tareas de minutos a horas sin supervisión.
https://t.co/DAw5f8nW7T

10. DeusData/codebase-memory-mcp
Knowledge graph del código para Claude Code, Cursor y Codex.
Se sincroniza solo con cada cambio. 100% local. Cero tokens extra.
https://t.co/Pog64Fu7vX

Junio de 2026 en GitHub está siendo una locura.

Guarda esto.

[Fact Sniffer @FactSniffer_]

@Osint613 What if Kim offerd to escort Iranian ships in case of a renowned US Naval Blockade?

[Fact Sniffer @FactSniffer_]

@MarioNawfal Well it wass free until they poke Iran

[Farea @FareaNFts]

Top 10 FREE API provider i found this week, in one place. 10 providers, zero credit cards 😳

frontier models (opus 4.8, gpt 5.5, glm 5.2, deepseek v4) for $0. benchmark + setup for each below

API KEYS (drop into cursor, cline, claude code, aider)

1. Runtime by Bad Theory Labs
10M free tokens/mo, one smart router across 340+ models
opus 4.8 / gpt 5.5 / deepseek v4 / glm 5.2 / kimi k2.6
> https://t.co/A2z7vO2Dr5 - google login, no card
> base url: https://t.co/4pf43QolqT - model: btl-2

2. Zenmux
glm 5.2 (62% swe-bench) / kimi k2.7 code (1T params) / step 3.7 flash
> https://t.co/lEbJrf6R0m - gmail, no card
> Models -> pick a free model -> create key -> copy base url

3. Mistral
1 billion free tokens on signup
mistral large 3 (77.6% swe-bench) / codestral / mathstral
> https://t.co/LCtl3gsPQC - gmail, no card
> base url: https://t.co/qyou43wATV

4. NVIDIA build
one key, 80+ models free
minimax m3 (#1 https://t.co/IO5FqUEAcS) / qwen 3.5 397b (77%) / kimi k2.6 / deepseek v4 flash
> https://t.co/0PDFmkcHDR - email + phone, no card
> base url: https://t.co/S9FYULI8nJ

5. Nex N2 (on OpenRouter)
397B frontier model, free on the :free endpoint, 80.8% swe-bench
> https://t.co/2OPz25RMQm - google login, no card
> base url: https://t.co/Z28Sw02Ch9 - model: nex-agi/nex-n2-pro:free

6. zcode (Zhipu)
3 million free tokens every day, resets daily
glm 5.2 (744B, MIT license, 1M context) built into their IDE
> download from https://t.co/OO4LPFEXZV - email, no card
> glm 5.2 is the default model

== FREE CHAT HUBS (no API key, just open and use) ==

7. Poe
daily free credits, refresh every 24h
claude / gpt / gemini / deepseek in one place
> https://t.co/DQqGbLxN0w - one login

8. Gumloop
5k free credits on signup
opus 4.8 / glm 5.2 / gpt 5.5
> https://t.co/ARbHOgNMfK - gmail

9. Agnes AI
free text + image + video in one login
replaces midjourney + runway
> https://t.co/qIbKZjb9oZ - no card

10. EvoMap
free claude / openai / gemini credits for verifying your github
> https://t.co/GlitVNTaJL - connect github + one public repo, credits land instantly

every API-key one is openai-compatible: works in cursor, cline, claude code, aider, hermes.

10 providers. $0. each would run $200+/mo if you paid.

bookmark this, claim as soon as possible bcz free tiers shift fast 👀

[Fact Sniffer @FactSniffer_]

@elonmusk @SurmountInvest India is doing exactly the same. Copy the India model of freebies distribution and you got it fixed in no time

[Fact Sniffer @FactSniffer_]

@gailalfaratx because critics give the best honest feedback to improve his products for free

[Fact Sniffer @FactSniffer_]

@Mrsinha Stop talking about them Sinha Ji. You are indirectly promoting and fueling their idea