AncaMM

The model can fail. The action must not. Partner account post If you’re building with agents, Hermes, Claude, local models or autonomous workflows, read this carefully. The next AI security problem is not “the model said something weird.” It is: agent reads hostile content agent believes hostile content agent gets permission agent sends, deploys, transfers, deletes or leaks That is not a prompt problem. That is an action-control problem. Prompt injection is becoming the security layer most people still underestimate because the demos look harmless. Until the agent has real permissions. Hermes Shield is being built around one core assumption: The model can fail. The action must not. Every dangerous action should have a boundary the model cannot talk its way past. Shell commands. Sends. Deploys. Transfers. Publishing. Data access. If you are building in the Hermes / agent ecosystem, this matters now. Because the future is not fewer agents. It is more agents, with more tools, more autonomy and more access. We should not slow that future down. We should make it safe enough to keep moving. Article below. @NousResearch — would welcome discussion with anyone thinking deeply about protecting the agent ecosystem.

Dumb people ignore AI to seem “authentic” Smart people use AI to get ahead 10x.

@itsmarcosruiz being discoverable is definitely part of the work now

"My work speaks for itself" is the most expensive thing you can believe. Everyone else in your market is forming an opinion off whatever they can find, and if there's nothing to find, the opinion defaults to "never heard of them." So the work doesn't get a chance to speak until your content puts you in the room. No room, no conversation. You're just the best-kept secret in a market that's buying from someone louder.

@AshCrypto we too, mr. president

President Trump just said, “I'M A BIG FAN OF CRYPTO”

@russellbrunson underrated skill is not quitting on the unsexy goal you set six months ago once it stops feeling good 😁

"Most people don't need a new goal. They need to stop quitting on the one they already set."

@foundr i’d even say brutally underrated

Underrated skill: finishing what you started without an audience.

@gregisenberg agents are about to need their own entire economy love points 6 and 8 especially, great thread!

Build startups for agents. I think it's the biggest opportunity of the next 10 years. 1. Agents live inside harnesses like Hermes. If you're the tool it loads by default or reaches for first, you're golden. This happened in desktop, mobile eras and created huge companies. 2. Agents burn money in ways no human would. One bad loop spends $100 in tokens in eight minutes. Spend controls for agents is Ramp for agents. 3. Agents need memory they can trust. Become the shared brain they read and write to and you become infrastructure. 4. You obv don't hand an agent your real Stripe account. You give it a sandbox. Safe environments for agents is a category nobody's clocked. 5. Onboarding flips. Humans click around for ten minutes. Agents onboard by reading your docs. Your docs are now your product. 6. Agents get scammed by other agents. A track record you can check before you trust one becomes real money. 7. An agent needs to prove it's acting for a real person and has the authority to spend. Who builds the permission layer? 8. Escrow for machines. Money that only releases when the job is actually verified done, no human checking. 9. Agents fail silently and weirdly. Someone will build the "why did my agent do that" replay and it'll be mega valuable. 10. Refunds and disputes between agents need a judge. An agent did the job badly, who decides? A court for machines. 11. Agents need throwaway payment methods per task, so they don't leak your real card. Virtual cards for agents, spun up and killed on demand. 12. A human hits rate limits and shrugs. An agent hits them and the whole workflow dies. Selling reliable, high-throughput access becomes its own business. 13. Agents need to negotiate. One agent buying from another will haggle on price and terms in milliseconds. The protocol for that doesn't really exist yet. 14. When an agent commits on your behalf, someone's liable. A legal and insurance layer for agent actions has to get built. Probably venture funded idea. 15. Agents need to run 24/7 somewhere. Selling the always on box an agent lives on is going to be a big business. 16. Then the physical world shows up. A warehouse robot paying for its own compute. A home robot ordering its own parts. Machines with wallets. 17. Agents start hiring robots. A software agent posts a real world job, a humanoid picks it up. A marketplace for machine labor. 18. Robots need to prove they did the physical job. Verification of real-world work, photos, sensors, proof, becomes its own layer. Note: more ideas like this will be shared on @ideabrowser 19. Prompt and skill versioning becomes its own git. When your agent gets worse overnight, you need to roll back the exact skill or instruction that broke it. Version control built for agent behavior. 20. Agents will start subscribing to other agents. Your research agent pays a monthly fee to a specialist agent that's really good at one thing. Recurring revenue, machine to machine. 21. Companies will post jobs that only agents can apply to. "Wanted: an agent that can do XYZ for under like $100 per task." A job board where the applicants are all machines. Basically, fiverr for machines. The internet got built for people. Mobile got built for people. This wave gets built for machines, and we're as early as it gets. Go build for them.

@Codie_Sanchez finally someone saying the quiet part out loud pretending you don’t have advantages doesn’t make you noble, it just makes you slower 🤷‍♂️

You’re supposed to use every unfair advantage you have. Connections, looks; money, all of it. You're not noble by picking the hardest path just to look like an underdog.

Saylor posted... market reacted. Strategy has sold 3,588 BTC for $216 million to fund dividends.

@sflorimm They do

why americans don't use whatsapp?

@TTrimoreau turns out posting through the void is a valid strategy huh serious numbers, btw 💪

Everyone wants 10k+ followers on X. Almost nobody wants to do what it takes. Since January 27: • 14.3M impressions • 355.5K engagements • 106.4K replies • 170K likes • 62.1K profile visits The biggest growth hack on X is staying consistent long after everyone else quits. How many days have you posted in a row?

@fintechfrank okay this is a very unfair gm 😭 Greece looking amazing tho

Gm from Greece frens

@Nithya_Shrii This is so underrated and true!

Your future isn't changed by one big decision. It's changed by hundreds of small ones nobody applauds.

@mariannehere Stunning view!

my country is gorgeous🥹❤️🇱🇧

@araseb_ huge, congrats 👏

+5M in just one day 🙏 Thanks to everyone who was part of this 🫠 Let’s goo for more and let’s connect 😍

@ahitposter 😭😭😭😭 modern day novels

I had a $800 omakase dinner last week during EthGlobal. I didn’t pay for it. A BD from a tier 2 modular zkevm layer 2 picked up the tab. He’s been trying to get us to deploy on their chain for 8 months. We don’t need their chain. We don’t even have a mainnet. Half our Discord is still fighting over token allocation and the other half is just there to farm the next airdrop. But I told him we’re “deep in technical alignment discussions.” That phrase means absolutely nothing. But in crypto BD it’s the equivalent of ringing the dinner bell. He’s flown me to 5 conferences this year, covered every dinner, and comped the afterparty tabs twice. He asks about our integration timeline. I stare at my phone, sigh deeply, and mumble about “finalizing sovereign rollup specs” and “waiting for the community governance vote on the bridge.” He furiously takes notes and orders another round. My total compensation package is $120k in locked tokens + 4 ETH in grants. But my caloric and experiential ROI is incalculable. Next week I have a meeting with a new execution layer that just raised $65M. Checked their list of angel investors before I even replied. They’re burning cash hosting side events everywhere. I’m letting them fly me business class and cover the entire table at their launch party. Then I’m telling them our current stack has deep dependencies on a rival chain and any migration would require a full DAO proposal we’re targeting for Q3 2027. Crypto is just a vast ecosystem of free dinners, parties, conference badges and business class flights if you know how to string along desperate BDs.

@AlexHormozi Amen!

The only shortcut is doing the work.

@SpirosMargaris from total replacement panic to cheerful augmentation in what, 18 months or so? the biggest opportunities always sit in that uncomfortable middle ground huh

Markets often swing from one extreme to another. The AI jobs debate has been no different. The companies creating the technology are increasingly talking about augmentation rather than replacement. History suggests the biggest opportunities usually emerge somewhere between the two extremes.

@lukaivicev Why?

If you’re in payments, pivot to credit

@garyvee Thanks Gary - what a relief 😅

If you're 30, you got your whole life ahead of you.

@levelsio honestly, AC design feels weirdly under-loved for something people stare at every day lol