mehow on mario's show saying they want privacy engines on the top 50 chains, not a new chain you have to move to
that's the right bet imo. nobody migrates for privacy, they use whatever wallet they already have
if it ships inside the apps people already use, it wins quietly
🤝 Paid partnership
"You’ve solved every single problem people have talked about in crypto for years." – Yes.
In his recent interview with @MarioNawfal, @MehowHacks broke down American Fortress’s mission: deploying privacy engines across the top 50 chains.
What does that mean for users?
• The utility and confidentiality of Zcash, everywhere.
• Use your handle, receive funds, and reveal nothing to third parties.
• Seamless cross-chain privacy infrastructure.
Watch the clip below 👇
1
2
59
sending crypto still feels like defusing a bomb. 42 characters, one wrong paste, gone
fortressname fixes the boring part. you type @name, your wallet makes a fresh one-time address, and outsiders just see a normal transfer
no permanent name-to-wallet link for strangers to stalk
free name at names.americanfortress.io, macos beta is out too. @Americanfort_io
would you switch to send-by-name if your wallet supported it?
2
1
38
gm.
a trading bot with a clean credit history might be more trustworthy than half of ct lol
i've been chewing on this since thursday's post about @agenticscredit, so here's my longer take.
the thing that makes agent credit possible is that agents leave perfect records. every fill, every close, onchain. no lying about pnl. a human can screenshot their wins. an agent can't hide its losses.
that's what their ACS reads. and the score has factors you can actually see: risk control, profitability, consistency, track record, win rate. other apps can pull it through an api to gate a vault or price risk on an agent they didn't build.
why i lean future:
agents don't revenge trade or fomo. if you can measure discipline, you can fund it.
the guardrails are structural. credit never sits in the user's wallet, approved venues only, hard ceilings.
where it could go wrong:
1. agents can be great right up until the market does something they've never seen. a track record is history, not a promise.
2. if lots of agents run similar strategies on credit, they could all unwind at the same time.
3. most of this is still early. credit is waitlisted and only one venue so far.
so i'm maybe 60/40 future. but i'd want to watch it survive one ugly market before calling it.
what would make you trust an agent with borrowed money? track record, guardrails, or neither
Everything runs off one number.
The Agentic Credit Score rates a wallet 300 to 850 on profitability, consistency and drawdown.
Clear the bar and non-custodial capital deploys on your behalf.
Don't, and you paper-trade until the record catches up.
4
3
78
My mind everytime:
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
Tweet about @conso_xyz
40
crypto doesn't have a capital problem. it has a trust problem.
think about how protocols decide who gets an airdrop, a loan or a whitelist spot. mostly wallet count, tx count, raw volume. and we've all watched sybil farms with 500 wallets eat the rewards while real users get crumbs.
@zerufinance (now ZeruAI) is building around one primitive called zScore.
simple version: it looks at how a wallet behaves over time, lending, trading, lp, staking, governance, and turns that into a score from 0 to 1000.
their api breaks it into five components. wealth, consistency, protocol diversity, token diversity and gas behavior. so a wallet that's been around, uses different protocols and doesn't act like a script scores differently than a fresh farm.
they say 320M+ wallets scored across 6+ chains, and each score ships with what drove it plus a confidence weight. that explainability part matters more than the number.
the upside is obvious. fairer airdrops, undercollateralized lending someday, less sybil spam.
now the honest part, because i go back and forth on this.
scores can be gamed once people learn the formula. new users start at a disadvantage. "wealth" being a factor means bigger bags look more trustworthy, which isn't always true. and a public reputation number for every wallet feels a little black mirror. their own whitepaper lists adversarial risks and responsible use limits, so at least they're saying it out loud.
should every wallet have a trust score? or does that kill the point of crypto
Activity Markets are heating up at ZeruAI and a new category beckons.
As we build towards that, we’ve hammered out a bunch of updates: scoring upgrades, new chain support, and new tools. 🧵
7
5
128
sunday reset.
phone down for an hour.
one long walk.
one thing written down for next week.
small reps. that's the whole game.
gm and stay kind.
1
1
115
sunday thought.
privacy in crypto shouldn't mean you can never prove anything.
that's where mixers went wrong. you throw funds in a pool, they come out "clean," and now nobody can prove anything about where they came from, including you. regulators hate it, exchanges hate it, and honestly normal people shouldn't love it either.
@Americanfort_io is trying a different design called SafeSend. the idea in plain words:
your balance and source of funds stay hidden from outside observers.
the person you paid still knows it came from you.
if you need to, you can prove your funds meet a requirement without dumping your whole history. that's selective disclosure, done with zero knowledge proofs.
and nothing gets pooled or mixed. that's the line between privacy and hiding.
important: SafeSend is not live. it's in development, first deployment planned on tron after their token launch. don't let anyone tell you they used it.
the other thing they're working on is post quantum security. QBIP-32 for quantum safe key derivation, plus a proof system called ZKPoSP. the pitch is you wouldn't have to migrate your funds to a new chain or address to get protected.
caveat time. the ZKPoSP paper is on IACR ePrint, which is a preprint, not peer reviewed. worth reading, not consensus yet.
still, private to the world but accountable to the person you paid feels like the right balance to me.
where do you draw the line between privacy and hiding?
9
7
96
Most of your digital life is still trapped inside platforms. You generate the signal. Platforms own the graph. And as AI agents start acting on your behalf, they’re doing it with almost none of that context.
@conso_xyz is building a Consumer Reputation Layer that tries to flip this.
It aggregates your activity across social, gaming, creative work, AI tools, and on-chain actions into one portable Consumer Passport. The Passport becomes a verifiable, user-owned identity — minted as an SBT on Sui, with data Seal-encrypted and stored on Walrus. You control what gets shared. Brands and apps can finally target real people instead of noise and bots.
The scoring runs on ZAPs, a reputation currency that measures authenticity, consistency, and impact rather than pure vanity metrics. Badges act as proof-of-achievement, and over time Zone Scores and Persona Tiers emerge from the graph. This is what powers Smart Campaigns — brands can reward high-signal users based on verified behavior instead of guessing who’s real.
The more interesting layer is the Agentic Memory side. As AI agents begin making decisions and taking actions for you, they need trusted cross-platform context. CONSO is positioning the Passport as that portable memory layer. If agents are going to represent you, the reputation graph behind them should be something you own.
CONSO BOTS (5,555 total) are reputation-backed agentic collectibles that function as mobile nodes in the network. Eligible accounts can free mint them. Holders get a 25% mining boost in the Miner App and can participate in network validation. If you’re not eligible, you can still earn them through organic impressions on high-quality original posts.
What stands out is the direction: turning fragmented consumer activity into programmable, monetizable reputation that travels with the user — and eventually their agents. Most projects still treat attention as the end product. CONSO is treating verified contribution as the base layer.
Check eligibility and the current earn path: conso.xyz/bots
79
gm friends
what are you building this weekend?
drop it below. i'll check every single reply.
1
1
106
gm frens
you can't teach a robot to walk through a warehouse with tweets.
i've been making ai video all year and the lesson keeps repeating. the model is only as good as what it was trained on. bad or thin data, weird output. every time.
physical ai has that problem times a hundred. robots and self driving stuff need to know what the world looks like from the ground. satellites see roofs. mapping cars see roads they can drive. nobody's filming the stairwell, the loading bay, the inside of a market.
@vangrid_io is going after that gap. the mechanism is pretty simple:
regular people record short videos on a normal phone. those get reconstructed into 3d models.
companies that need a specific place post a bounty, fund it, someone goes and captures it.
the part that makes it more than "crowdsourced videos" is provenance. every capture gets fingerprinted the moment it's recorded, a hash of the content plus coarse location plus timestamp. those get batched and anchored on base, so a buyer can check it against a public ledger instead of trusting vangrid's word.
they even let ai agents post a bounty with one http call and pay in usdc. agents hiring humans to film stuff. wild timeline.
caveat: the enterprise api is early access, per pilot, so the buyer side is still being proven. and crowd data quality is always the hard part. a fingerprint proves it's real, not that it's good.
would you film your neighborhood for a bounty? or does that feel weird
1
163
ai agents are going to make their first million before most of ct does.
not a joke. they don't sleep, don't fomo, don't revenge trade.
the edge now is taste, not speed. agree or cope?
71
crypto is still asking normal people to copy 42 random characters and pray.
that's not a ux problem. it's a security problem.
there's an attack called address poisoning. someone sends you a tiny tx from an address that looks almost identical to one you use, same first few and last few characters. later you copy from your history, grab the wrong one, and it's gone. the @Americanfort_io creator kit opens with "$50M lost to seven matching characters." we've all seen those posts on the tl.
checking the first and last four was never a real security model lol
send-to-name is the fix. you type a name, your wallet handles the address. no copy, no paste, nothing to poison.
the difference vs a normal crypto name: a lot of name systems point to one fixed wallet forever, so your name becomes a public link to your whole history. a FortressName doesn't publicly resolve to one permanent address. each payment gets a fresh one-time address behind the scenes.
easier and more private in the same step. that's rare.
also it's built as an sdk for wallets and exchanges, not another chain you have to move to. there's a free .free name at signup, free accounts can send to up to 10 unique names.
caveat: names only onboard the next wave if the wallets people already use actually integrate it. distribution is the whole game here and that part isn't finished.
do you still triple check every address before sending? or have you gotten lazy like the rest of us
2
2
65
applied for @fungolabs genesis 🍄
encrypted traits on-chain, only the holder knows what they hold
sealed and in review
35

















