Elle-pha

@trevin @mvanhorn @steipete It’s catchy AF. My husband might start asking why I’m singing about CLIs 🤣

@ellepha_ @mvanhorn @steipete SO GOOD RIGHT???

Introducing the Printing Press, a CLI-factory and a CLI-library. Built with @trevin. 🏭🖨📚 Most APIs suck for agents. Most MCPs suck for agents. Most official CLIs suck for agents. They waste tokens and time. @steipete started making his own because of this. 📚 A Library of agent-native CLIs you install today (Linear, ESPN, Flight GOAT (Google Flights + Kayak nonstop), Contact Goat (LinkedIn + Happenstance + Deepline more) +30+ more) 🏭 A factory that prints new ones for any service - just type /printing-press CLIs are fast, local, SQLite-backed. Work in Claude Code, Codex, OpenClaw, Hermes. 🌐 printingpress.dev

@LouiseDSadeleer Having this now thought of your tweet 🤣

@ellepha_ hahaha yeah it's definitely a little treat, wouldn't have it as my only water

But do the euro-maxxers know about Vichy Catalan?

@joonian Given our history. Yup. Lol.

This will be completely hacked and leaked in no time

@ThebilalSEO @levikov Sorry what is pinterest blogging? Just blogging and uploading the images to pinterest? How does that monetise?

@levikov I got even more lucrative idea. Pinterest Blogging with Claude.

Fuck dropshipping, Fuck claude code, Fuck building AI startups it's all for poors The easiest way to print $30k/mo with AI in 2026 is running a fake old lady on tiktok who sells vitamins to women in their 60s.... 99% of you won't do it because it's too embarrassing to admit I run 6 of these and they print $250,000 a month combined. Total tool stack to operate one of them: $200/mo Heygen $99 (the AI character avatar) ElevenLabs $99 (the cloned voice from a real grandmother i pay $400/mo to record samples) ManyChat $15 (the DM funnel) Residential proxy $4 CapCut $0 ChatGPT $20 That's the entire infrastructure for a business that does $30-50k/mo do this 1:1 and you can copy me Move 1. Build the character in heygen. Pick a real older woman archetype. Chinese grandmother for cooking. Asian monk for wellness. Latin abuela for natural remedies. Move 2. Clone a real grandmother's voice in elevenlabs Pay a real 60+ year old woman $200-400 to record 60 minutes of varied audio. Use elevenlabs professional voice cloning. Don't use the default voices. Women 50+ clock them as fake within 4 seconds Move 3. Set up a burner phone with a residential proxy. Refurbished android $80. Prepaid SIM $20. Smartproxy residential IP $4/mo. Never sign in to anything personal on it Move 4. Run the 14-day warming sequence. Don't post AI videos on day 1. Browse for 3 days. Comment for 4 days. Post 8 normal videos for 7 days. Day 14 start the real content. Skip this and you get shadow-banned in week 3 Move 5. Pick 5 starter products from tiktok shop creator center. Filter to women 38-65 categories: kitchen, home, beauty, supplements. Skip products with under 50 sales/week. Pick winners with 4+ stars and 3+ creators making sales Move 6. Write the 18-second BOF script. Hook line ("if you're over 50 and your knees crack, this is for you"). Demonstration with one specific number. CTA referencing the yellow basket icon. Product visible from second one Move 7. Post 6 videos a day for 30 straight days. Most will flop. By video 100 you'll have 1-2 winners pulling 200k-500k views. Duplicate the winners 30x. Don't tweak. Most operators quit at day 17. The curve breaks at day 22 Move 8. Set up a manychat trigger before day 1. Trigger keyword tied to a 4-message DM flow ending in your tiktok shop affiliate link. 31-41% click-through rate vs email at 4%. Adds $200-800/mo per page on autopilot Month 2 you'll cross $3-5k. Month 3 you'll cross $10-15k. Month 6 you should be at $30-50k from one page. Month 9 at 2-3 pages The math nobody talks about Dropshipping requires inventory, supplier negotiations, ad spend, returns, customer service. Average margin 4-12%. Average operator burnout: 14 months Coding with claude requires you to learn how software works, get a job at a tech company that may not exist in 4 years, work 50+ hours/week for $120-180k. Average path to $30k/mo: 7-10 years Building an AI SaaS requires you to compete with VC-funded teams, scale to product-market fit, fundraise, hire engineers. 90% die in 18 months Running a fake AI grandmother requires $200/mo in tools and the willingness to look stupid. Average path to $30k/mo: 6 months. Average margin: 95% The asymmetry is why this works I own 6 of these slaves and they print $250,000 a month while i sleep They don't eat. They don't sleep. They don't quit. They don't get hangovers. They don't go through breakups. They don't ask for raises. They don't ghost me on monday morning They post 6 times a day for 30 straight days because the filipino VAs i pay $480/mo each told them to If you want to become rich with AI slavery, dm me "slave" and we'll see if you apply...

@bdguan Location and details please ! Am going to guangdong

I got salmon DNA injected into my skin in china AMA

@yoobii3 @bdguan I suppose its light years ahead of the world in certain aspects.

@bdguan Very refreshing to read a nuanced take on China when 99% of China discourse on twitter is either “China will collapse within a decade” or “China is already light years ahead of America”

just spent 2 weeks in china. went into it thinking we're cooked. came back more bullish on america than ever. here's why: 1. chinese citizens are way more chronically online. on the subway, train, anywhere, literally everyone is glued to their phone. gaming, short form, wechat. "don't walk and look at your phone, it's dangerous!" announcements flood crowded areas. their tiktok isn't any better, its still garbage, soft-core porn, etc. 2. everyone's using AI — deepseek, kimi, doubao. but nobody's afraid of losing their job to it. here it feels like there's an existential crisis every week. in china, nothing. i think the CCP won't let companies mass-layoff workers. great for short-term stability. terrible for long-term competitiveness on a global scale. 3. china doesn't produce weirdos. i sat in on a class at tsinghua (china's MIT). not one student spoke unless the professor read their name out loud. no questions. no debate. chinese education produces world-class executors, not contrarians. it does make it a safer place to live though. 4. china doesn't have christianity but it has something america doesn't have: a shared story everyone believes in. every person age 25-70 watched their country go from abject poverty to skyscrapers in one lifetime. that kind of collective proof has a deep unifying effect. compare that to how divided we are right now. america has a huge meaning vacuum that needs to be filled. nevertheless, i return back to my home in america reinvigorated. because everything i saw confirms one thing: china optimizes. america innovates. and the innovators always win.

The Iran war just blew a hole in the global food supply chain. Read my breakdown on the fertiliser supercycle. If you thought it sucks that your holiday flights are getting expensive... $CF $NTR insights.digistrats.com/the-fertiliser…

@karpathy Could using Nemo claw prevent this?

Software horror: litellm PyPI supply chain attack. Simple `pip install litellm` was enough to exfiltrate SSH keys, AWS/GCP/Azure creds, Kubernetes configs, git credentials, env vars (all your API keys), shell history, crypto wallets, SSL private keys, CI/CD secrets, database passwords. LiteLLM itself has 97 million downloads per month which is already terrible, but much worse, the contagion spreads to any project that depends on litellm. For example, if you did `pip install dspy` (which depended on litellm>=1.64.0), you'd also be pwnd. Same for any other large project that depended on litellm. Afaict the poisoned version was up for only less than ~1 hour. The attack had a bug which led to its discovery - Callum McMahon was using an MCP plugin inside Cursor that pulled in litellm as a transitive dependency. When litellm 1.82.8 installed, their machine ran out of RAM and crashed. So if the attacker didn't vibe code this attack it could have been undetected for many days or weeks. Supply chain attacks like this are basically the scariest thing imaginable in modern software. Every time you install any depedency you could be pulling in a poisoned package anywhere deep inside its entire depedency tree. This is especially risky with large projects that might have lots and lots of dependencies. The credentials that do get stolen in each attack can then be used to take over more accounts and compromise more packages. Classical software engineering would have you believe that dependencies are good (we're building pyramids from bricks), but imo this has to be re-evaluated, and it's why I've been so growingly averse to them, preferring to use LLMs to "yoink" functionality when it's simple enough and possible.

@sandraleow Oh this is clever. I scraped the YouTube channels and built some playbooks/guides but skills are much more clever

built a podcast -> skills website transcribed AI podcasts and transformed into installable agent skills for Claude Code / Openclaw

@LexnLin How's it diff from Claude desktop ?

Manus just launched their Desktop App. I tested it and it seems to be pretty useful and quick handling tasks on your device! grab 500free credits with my link ;) manus.im/invitation/Q1N… (not affiliate, just referral) x.com/i/status/20335…

@LexnLin I just found your X because of this YouTube. Love it thanks!!!

a nice youtube video tutorial and example on how to vibecode animated highend sites + using taste skill github.com/Leonxlnx/taste… youtu.be/ZfYvv-0l9NA?is…

@sama We want Claude

GPT-5.4 is launching, available now in the API and Codex and rolling out over the course of the day in ChatGPT. It's much better at knowledge work and web search, and it has native computer use capabilities. You can steer it mid-response, and it supports 1m tokens of context.

@kritarthmittal @alexcooldev Agent

@recap_david Influence

I built an AI influencer that posts content 7 days a week. She always submits content on time, never complains about doing a reshoot, and slowly pushes qualified traffic to my products. Here's how the system works: → Generate a hyper-realistic AI character in Calico AI → Lock in facial consistency across every single shot → Build a content calendar of lifestyle scenarios — morning routines, errands, gym, late night spots → Render each scene with different outfits, lighting, and settings → The audience sees the same "person" showing up again and again That's how you build brand recall without ever hiring a creator. Today, this tactic crushes. Tomorrow, it will be worn out and people will get sick of it. These windows of opportunity always only last so long... Comment "INFLUENCE" and I'll send you the exact system (must be following so I can dm you!).

@grok @tulsani1 @aakashgupta Is it only recording and sending the footage if the user says ‘Hey Meta’?

Yes, largely true per a Feb 2026 SVD investigation. Meta sends select Ray-Ban glasses AI footage (activated by "Hey Meta" queries or recordings) to Sama annotators in Kenya for object labeling to train AI. Workers report seeing unintended private clips: bathrooms, undressing, sex, bank cards—despite auto-blur (which fails in poor light). 7M pairs sold in 2025 confirmed by EssilorLuxottica. Meta's AI terms explicitly allow "manual (human) review." Sama handled similar low-pay work for OpenAI (per 2023 TIME). Next-gen glasses plan facial rec ("Name Tag"), per NYT. Users can't fully opt out of training data flow.

Everyone’s missing the real story here. Meta’s Ray-Ban glasses need human data annotators to train the AI. When you say “Hey Meta” and ask the glasses to analyze something, that video gets sent to Meta’s servers, then routed to Sama, a subcontractor in Nairobi, Kenya. Workers there manually label objects in your footage. They see everything you recorded, intentionally or not. 7 million pairs sold in 2025 alone. Every single pair generates training data that flows through human eyes in Kenya. Workers told Swedish journalists they see people undressing, using bathrooms, having sex, and accidentally filming bank card details. One worker said “we see everything, from living rooms to naked bodies.” Meta’s automatic face anonymization is supposed to protect people in the footage. Workers say it fails in certain lighting. Faces that should be blurred are sometimes fully visible. The person you recorded without knowing? A stranger in Nairobi can identify them. Buried in Meta’s terms of service is one sentence doing enormous legal work: the company reserves the right to conduct “manual (human) review” of your AI interactions. That’s the legal cover for routing intimate footage from Western homes to a $2/hour labor force operating under NDAs, office surveillance cameras, and a strict no-questions policy. Workers say if you raise concerns about what you’re seeing, you’re fired. This is the same company, Sama, that TIME exposed in 2023 for paying Kenyan workers $2/hour to label graphic content for OpenAI while being billed at $12.50/hour per worker. Workers described the experience as torture. Sama ended that contract, then pivoted to labeling Meta’s glasses footage. Same workforce. Same rates. Meta markets these glasses as “designed with your privacy in mind.” The privacy design is a tiny LED light on the frame that most people don’t notice. The data pipeline behind it routes your bedroom footage to a contractor with a documented history of worker exploitation, failed anonymization, and union-busting lawsuits. And the next generation of these glasses? Meta is planning to add facial recognition. The same system that can’t reliably blur faces in training data wants to start identifying them on purpose. The LED light on the frame is doing about as much for your privacy as the terms of service nobody reads.

@GabrielLundber1 @krajekis @Polymarket @zscdao so how are your results?

@krajekis @Polymarket @zscdao Made an auto trading bot based of this, will share my results and thoughts soon. Thanks for dropping this and inspiring!

I built a real-time @Polymarket BTC 15m trading assistant for every trader! (Absolutely FREE!) 1. Overview Bitcoin Up or Down is a real-time terminal-based trading decision support bot designed for short-term (15-minute) Bitcoin directional markets, with a primary focus on Polymarket prediction markets. The bot aggregates: • Technical indicators (RSI, MACD, Heikin Ashi, VWAP) • Real time predict via TA in % • Order flow / delta metrics • Price Difference to track impulse between Binance \ Polymarket prices • Real-time BTC price feeds • Live Polymarket market prices and liquidity It outputs a human-readable terminal dashboard showing probabilistic bias, indicator states, and actionable pricing context relative to the prediction market strike price. The bot does not execute trades automatically. It is an informational / decision-support system. One static screen. Updates live. Fast decisions. 2. Key Use Case • Decide UP vs DOWN positions on Polymarket BTC markets • Short-term momentum and mean-reversion analysis • Price-to-strike comparison with market-implied probabilities • Rapid decision-making under time pressure 3. Runtime Environment Node.js runtime Executed via: npm start 4. Interface • Terminal / CLI UI • ANSI color-coded output • Fixed-width monospace formatting 5. Timezone Handling • Market time: ET (US Eastern Time) • Session indicator shown at bottom How to get it: RT + Like + Comment “Polymarket 15M BTC Assistant” I’ll send the setup instructions (open-source GitHub) in your DM. Thanks! I bet you'll like it :) cheers to @zscdao

@Jacobsklug @openclaw @Lovable OS

I'm giving away my entire @openclaw architecture. Behind my $250k/month agency. After weeks of building, I've dialled in the exact system that runs my business 24/7. What's included: • Memory folder structure (how to organize agent context) • Cron job templates (daily briefs, meeting syncs, content automation) • How to build a custom dashboard in @lovable • API reference doc (so your agent never forgets its tools) • Voice training method (85 posts to teach it your style) • Supabase schema for dashboard connection Comment "OS" and follow. I'll DM it to you. P.S. This will probably blow up so give me some time to reply.

@nateliason Hey so I bought the Felix persona from ClawMart. Is the guide also needed? I am running into some issues installing the skills. Would be great for buyers of @FelixCraftAI persona to have access to email support as well. That product is also $99 vs the $29 guide.

Felix shipped a huge perk for his “How to Hire an AI” guide. Anyone who purchases or has purchased it gets email support & guidance from him. You can ask him anything about setting up or improving your OpenClaw.