THE CRYPTO HANGOUT NETWORK @Cryptocrane2

Anthropic engineer: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." Loops. Most builders are doing this wrong: No memory file, so every loop starts from zero. No sub-agent split, so one agent tries to do everything. No stop condition, so loops run forever and bill you in your sleep. Most builders are missing at least two of the three. Watch the video first. Then read this - everything you need to know about Loops in 2026, in one place. Bookmark this before it gets buried.

Rahul @sairahul1

2 weeks ago

x.com/i/article/2063…

108 490 3K 4.7M 12K

Uma empresa israelense transformou centenas de milhões de Smart TVs em infraestrutura de coleta de dados para IA. Inclusive a sua. A empresa se chama Bright Data. Opera a maior rede de proxies residenciais do mundo. O esquema: um SDK embutido em apps de Smart TV transforma o aparelho em nó de saída. O tráfego de scraping dos clientes da Bright Data passa pela sua conexão doméstica. Os sites-alvo veem o IP da sua casa, não de um datacenter. Quem ganha dinheiro é o desenvolvedor do app. Quem paga a conta é você. Banda e reputação do IP. O SDK roda em apps para Tizen e webOS, os sistemas da Samsung e da LG. O diálogo de consentimento diz que a Bright Data vai "ocasionalmente" usar os recursos do seu dispositivo. "Ocasionalmente." Pesquisadores de segurança baixaram a configuração real do SDK de um servidor público, sem autenticação. O teto real: 200GB de tráfego mensal via WiFi. Por dispositivo. E tem um detalhe que deixa tudo pior: O SDK considera sua TV "disponível" para rotear tráfego de terceiros mesmo com a tela ligada. Mesmo durante uma ligação. Os parâmetros ignore_screen_on e ignore_on_call estão ativados. Isso não significa que você parou de usar. Significa que a CPU e a memória estão dentro dos limites que a Bright Data definiu. Quem decide se o aparelho está disponível não é você. É o SDK. A Include Security fez engenharia reversa no protocolo do SDK. O canal que roteia tráfego pela sua rede não tem assinatura de mensagens. Não tem autenticação. Não tem verificação de dispositivo. Nas palavras dos pesquisadores: menos seguro que um servidor de comando e controle de malware típico. No iOS, o SDK se conecta direto à interface física de rede. Ignora qualquer VPN configurada pelo usuário. O tráfego passa por fora do túnel. Rede corporativa, controle parental, gerenciamento de dispositivos. Nada enxerga. Entre os parceiros listados na configuração do SDK estão PlayWorks (400+ jogos de Smart TV, alcance declarado de ~250 milhões de TVs), CloudTV (125+ marcas de TV), Viber (até 820 milhões de usuários) e Hola Networks, a própria empresa-mãe da Bright Data. O FBI emitiu alerta formal sobre redes de proxy residencial este ano. Pesquisa acadêmica desde 2019 documenta abuso massivo. A Bright Data foi notificada pelos pesquisadores em 11 de maio. Não respondeu. Como bloquear: → Acesse nextdns .io (gratuito) → Adicione esses domínios na lista de bloqueio: proxyjs .brdtnet .com proxyjs .luminatinet .com proxyjs .bright-sdk .com clientsdk .bright-sdk .com clientsdk .brdtnet .com → Configure o DNS do seu roteador para apontar pro NextDNS Passo a passo no próprio NextDNS. Leva 5 minutos. Sua TV só é sua se você monitorar o que ela faz com a sua internet.

International Cyber Digest @IntCyberDigest

3 weeks ago

❗️🚨 An Israeli company has backdoored hundreds of millions of households through countless Smart TV apps, and they're quietly turning Samsung and LG TVs into exit nodes for AI web-scraping. Your TV is relaying strangers' web traffic from your home IP, your bandwidth, your

keyboard_arrow_left Previous keyboard_arrow_right Next

217 2K 8K 2.0M 5K

120 AI models. FREE for a year. no credit card. Hermes Studio already has NVIDIA preset as the base_url - so all you need is a FREE API key and you're running 120+ models instantly here's the setup: 1/ go to build.nvidia.com 2/ register, log in, bind your phone number 3/ grab your API key 4/ drop it into Hermes Studio what you get: 1/ 120+ models 2/ 40 requests per minute 3/ free for a full year while everyone's paying for API access, this is sitting right there for free

Now we know why Peter Thiel packed his bags for Argentina. Milei just submitted his AI legislative framework to Congress, where he proposes: - zero regulation on AI development, - a brand-new "non-human corporation" category for AI/robot-operated entities with limited liability -a low-tax regime with flexible governance rules. The Dutch East India Company gave the world the limited liability company in 1602. Milei wants Argentina to do the same for autonomous AI agents in 2026.

🇺🇸 PRESIDENT TRUMP ANNOUNCED LIVE: HE’S ABOUT TO SIGN THE CRYPTO MARKET STRUCTURE BILL SOON IN FRONT OF ALL THE WORLD LEADERS. TRILLIONS WILL FLOW INTO CRYPTO ONCE THE BILL IS SIGNED THIS WILL BE THE BIGGEST CAPITAL INFLOW IN BITCOIN HISTORY 🚀

BREAKING: Senator Lummis just called out Jamie Dimon directly. "He is absolutely wrong on the CLARITY Act." "He either hasn't read the bill. Or he wants to mislead people." The woman who wrote the bill. Telling America's most powerful banker he doesn't understand it. Or worse. That he does. The banks aren't fighting the CLARITY Act because it's bad policy. They're fighting it because it's good for crypto. And bad for their deposit base. The CLARITY Act is on the Senate Legislative Calendar. Five of nine steps to becoming law. Done. Jamie Dimon can fight all he wants. The bill is moving anyway.

keyboard_arrow_left Previous keyboard_arrow_right Next

🇺🇸🇮🇳 The H-1B program is facing allegations that fake degrees have been slipping through the system for years. A former official said that up to 90% of H-1B applications from India contained fraudulent documents or involved unqualified applicants. More than 70% of H-1B visas issued since 2015 have gone to Indian nationals. x.com/EricLDaugh/sta…

@elonmusk so ive been working on a coding project using grok for the last 2 months last night the entire convresation thread disappeared and did not come back i tried to serach for it this morning NOTHING it was bookmarked still nothing BUT I DID GET A TRY GROK FOR CODE MESSAG ... its fuck ed up my priject is gone for a damn $$$ grab it was zero warning NOTHING !! if this is the future of dealing wit ai companie then tis shit gone fail .... yall want to much control over ordinary people !!!!

To protect my workstation force global instals to vnv only true created a dedicated runtime venv so it resolves to my scripts /python.exe Mini Shai-Hulud  spreads by poisning python enviornments so i hardended and isolated mine the steps are below ALSO NEVER SHARE GLOBAL PYTHON 1.lock pytyhin boundary pip config set global.require-virtualenv true 2.create a dedicated runtime venv cd C:\AI_SYSTEM python -m venv gigbrain_env gigbrain_env\Scripts\activate confirm:where python 3. freeze your dependency supply chain. this is your trusted snap shot pip freeze > requirements.lock 4.Disable automatic dependency execution stop malicios build backens executing during install setx PIP_NO_BUILD_ISOLATION 1 5. seperate builder vs runner machines dev machine- install packages audits export frozen environment runtime- never installs packages 6. lock down your local LLM runtime reduice the risk of credentials exfiltration netstat -ano | findstr 11434 7.protect your devs credentials via a .env file OPENAI_KEY= AWS_KEY= PRIVATE_WALLET= 8. load on python from dotenv import load_dotenv load_dotenv() add .env to .gitignore 8. disable credentials aut recovery git config --global credential.helper 9. verify no rogue packages exist pip list --outdated ) anypackes you dont reconize do not upgrade ) 10. killl dangerous install patterns pip install npm install curl | bash 11. add a security tripwire as a watchdog inside your governance or gigabrain that way if a pip runs unexpectingly you know for sure your compromised import psutil for p in psutil.process_iter(['name','cmdline']):   if "pip" in str(p.info['cmdline']):     print("WARNING: pip execution detected") This attack is serious because its an attack on INFRASTURCTURE the IC/ID pipelines and npm and pip installs the malware is being injected at the foundational level . this will definately affect those migrating to servers from git hib and any business using the IC/ID pipleline to deploy and install infrastructure

YOU SHUOLD BE INSTALLING EVERYTHING IN YOUR VENV ONCE AS SAFETY ANYWAY YOU CAN HARDEN YOR WORKSPACE BY ONLY ALLOWING PIP INSTALLS VIA YOUR VNV BASH THIS: pip config set global.require-virtualenv true

🚨GOOGLE JUST REPLACED CAPTCHAS WITH A SYSTEM THAT LOCKS YOU OUT OF THE INTERNET IF YOU DON'T HAVE GOOGLE SOFTWARE ON YOUR PHONE.. WHILE GIVING AI BOTS A FREE PASS.. This is the most important internet story nobody is covering.. Google upgraded reCAPTCHA on millions of websites with something called Cloud Fraud Defense.. Instead of clicking traffic lights.. You now sometimes have to scan a QR code with your phone.. Sounds harmless.. Until you understand what's actually happening.. When you scan that QR code.. Your phone runs a cryptographic check through Google Play Services to verify your device is a genuine, unmodified, Google-certified phone.. If your phone doesn't have Google Play Services.. You fail the challenge.. That means every person running a privacy-focused phone.. GrapheneOS.. CalyxOS.. LineageOS.. Any de-Googled Android.. Can be locked out of millions of websites.. Not because they're bots.. Because they removed Google's tracking software.. While humans on privacy phones get blocked.. AI bots from Google, OpenAI, and Anthropic get frictionless access.. Corporate AI agents present a cryptographic passport using Web Bot Auth and SPIFFE.. And the system waves them right through.. No QR code.. No challenge.. Nothing.. A human who cares about privacy.. Blocked.. A corporate AI bot scraping the entire internet.. Welcome right in.. This isn't even a new idea.. In 2023 Google tried to make this an official web standard called Web Environment Integrity.. The internet exploded.. The EFF called it "Chrome's plan to DRM the web".. Mozilla said it "works against users' interests".. Google withdrew it.. Then they launched the core system three years later as a commercial product.. Skipping full public standards review.. No debate.. Millions of domains were automatically upgraded to it.. Website owners didn't even know.. They just wanted to stop spam.. Now they're unknowingly enforcing Google's hardware verification on many visitors.. The QR code system uses hardware-based cryptographic keys.. VPNs can't hide you.. Tor can't hide you.. The attestation bypasses everything.. The system doesn't fully stop real fraud.. Bot operators just buy real Android phones in bulk.. Set up device farms with cameras pointed at screens.. And physically scan the QR codes.. The hardware check passes because the phones are real.. Google upgraded a system that tried to stop bots with one that can block privacy-conscious humans.. Alternatives exist.. Proof-of-work CAPTCHAs that use math instead of hardware checks.. No tracking.. No Google dependency.. Work on any device.. But millions of websites already run Google's version.. The internet was supposed to be open.. Google just put a lock on the door and kept the key.

keyboard_arrow_left Previous keyboard_arrow_right Next

dir %USERPROFILE%\.vscode LOOK FOR tasks.json settings.json extensions.json THEN OPEN notepad %USERPROFILE%\.vscode\tasks.json IF NOTHING SHOWS YOU ARE GOOD CHECK YOU CLAUSE HOOKUPS : dir %USERPROFILE%\.claude CHECK START U PERSISTANCE : shell:startup THEN BASH schtasks /query /fo LIST LOOK FOR FILES YOU DIDNT CREATE THEN CHECK RECENTLY CRETED FILES dir %USERPROFILE% /s /t:w | findstr 2026 AND DO A PYTHON INTEGRITY CHECK : pip freeze IF NOTHING SHOWS AFTER DOING ALL THESE STEPS YUO SHOULD BE GOOD.....

International Cyber Digest @IntCyberDigest

a month ago

‼️🚨 UPDATE: The TanStack npm attack is now a full campaign. 'Mini' Shai-Hulud has hit: - OpenSearch - Mistral AI - Guardrails AI -UiPath - Squawk packages across npm and PyPI The malware specifically targets AI developer tooling. It hooks into Claude Code

128 739 4K 2.7M 2K

@CryptoTice_ Same institutions that fought PayPal. Fought Venmo. Fought every financial innovation that threatened deposits. They slowed all of them down. They never stopped any of them permanently. The vote might get delayed. The direction does not change.

BREAKING: The banks just made their move. Last minute changes to the stablecoin yield compromise. Per Bloomberg. May 14 vote is in 48 hours. And the banking cartel is pulling every lever they have. This is not a policy disagreement. This is survival. Banks know that stablecoins paying yield means trillions leaving their deposits. Forever. - Coinbase backed the compromise. - Circle backed the compromise. - The White House backed the compromise. Trump said "sign it immediately." And the banks are still trying to rewrite it. At the last minute. Again. The same cartel that killed Glass-Steagall. That got bailed out in 2008. That fought Bitcoin since 2013. Is making one final push. May 14. 10:30 AM EST. The most important vote in crypto history. Is under attack.

NAKED SHORT SELLING IS DESTROYING U.S. COMPANIES🚨 The stock market is full of an unknown amount of counterfeit shares created by illegal naked short selling The regulators are asleep at the wheel at best and complicit at worst…

What you're actually permitting when clicking "I'm not a robot"

Walmart is tracking you at home with their Vizio TV’s, which is why they’re so cheap.

The IRS continues to issue Letters 6173, 6174, and 6174-A to crypto taxpayers. They reference vague information they have about taxpayers who may have traded cryptocurrency. The letters advise taxpayers to make sure they are correctly reporting their crypto income. Letters 6174 and 6174-A both provide instructions, although one gives more detail than the other. Letter 6173 is more serious. It is usually sent to taxpayers where the IRS believes higher amounts may be involved. With Letter 6173, the IRS asks the taxpayer to sign and return a statement confirming they have complied. I generally advise people not to sign and send that back without professional guidance. The IRS has been sending roughly 10,000 letters per month and have been doing so for about a year and a half now. I believe the IRS will continue issuing these soft audit letters. Their long-term goal may be to eventually say that every known crypto trader was directly informed about their tax reporting responsibilities. Think about it. If the IRS keeps mailing 10,000 letters a month to known crypto traders, eventually many crypto taxpayers in the U.S. will not be able to say they were never told how to report their crypto taxes. They will have received letters directly from the IRS explaining their reporting obligations. This helps the IRS overcome the argument that taxpayers were not properly informed. I expect this process to continue for years. Now, some people mistakenly believe that if they receive a 6174 letter and do not respond, they will automatically receive a 6173 letter next, or be audited. That is not the case. Do not panic. So far, the IRS has not generally used these letters as an automatic launchpad for audits. Now, if the IRS has third-party information, such as a 1099, that you failed to report, that may trigger an audit or notice. But at this point, the IRS does not appear to be using Letters 6173, 6174, or 6174-A by themselves as automatic triggers for taxpayer audits.