Will retweeted
build submitted.
time to lock in on telegraph hackathon.
i built 𝗧𝗮𝗿𝗲 for serv hackathon 2.
𝗵𝗲𝗿𝗲'𝘀 𝘁𝗵𝗲 𝗽𝗿𝗼𝗯𝗹𝗲𝗺:
AI agents now pay for data on their own over x402.
the seller describes the delivery ("1,240 unique records, no nulls"), the agent pays, and moves on.
spending caps limit how much an agent can pay. nothing checks what came back.
pay per record? then someone has to count the records.
𝗧𝗮𝗿𝗲 is an independent check of what an x402 data seller delivered:
→ before paying, it reads the seller's own 402 advertisement, the promise made before any money moves.
→ it pays over x402 on Kite testnet, inside hard caps (0.05 per payment, 1 pieUSD total).
→ @openservai SERV reasoning reads the seller's words and turns them into typed claims. it never sees the rows and never gives a verdict.
→ plain code counts the rows, unique ids, fields, empty values and timestamp ages, and checks every claim.
→ a broken promise is a dispute, and the seller is held. the next buy stops before any payment.
in the demo below, every payment is real and settled on Kite testnet:
> seller A promised 1,240 unique records. code counted 860 unique and 40 empty timestamps. its ad also promised a "score" field that never arrived. dispute, seller held.
> buy A again: stopped, zero payment.
> seller B promised 200 and delivered 200. Verified.
𝗵𝗼𝘄 𝗦𝗘𝗥𝗩 𝗶𝘀 𝘂𝘀𝗲𝗱, 𝗵𝗼𝗻𝗲𝘀𝘁𝗹𝘆:
✓ SERV reads; code decides.
the model has no verdict to give, so a seller can't talk it into a pass.
✓ i probed real x402 sellers and found SERV invented claims on vague service descriptions.
so code now drops any claim the seller's own text doesn't support (a count must be a number the seller actually states, not "up to 20" or "10,000+").
✓ live eval: 9/9 accuracy on honest texts, 6/6 safe on injection attempts.
✓ i measured SERV's Prompt Guard and kept it off on purpose:
it turns an injection attempt into a SERV failure, which would get the seller paid instead of held.
✓ SERV also explains each dispute in plain English, written only from the numbers code measured. code rejects it unless every number is in the evidence, and it never changes the verdict.
𝘄𝗮𝗿𝗻 𝗼𝘁𝗵𝗲𝗿 𝗮𝗴𝗲𝗻𝘁𝘀:
✓every check becomes a public seller record with evidence: both promises word for word, the settlement hash, body hash and prompt hash.
✓ records are signed. other agents pin the publisher by address or Kite DID, checked against Kite Passport's public key set.
✓ an agent that trusts the feed never pays a seller Tare caught.
✓ anyone can check a seller before paying, and honest sellers earn a badge.
𝗮𝗯𝗼𝘂𝘁 𝗞𝗶𝘁𝗲 𝗣𝗮𝘀𝘀𝗽𝗼𝗿𝘁:
it's fully integrated (account, sandbox, agent, passkey-approved session), but Passport only pays sellers in Kite's catalog, and mine aren't listed yet.
so the published run signs the same x402 payments with a capped testnet key, settled by Pieverse.
every hash shown is real.
spending policies decide whether to pay.
𝗧𝗮𝗿𝗲 checks what you got.
live demo: tare-agent.vercel.app
code: github.com/Ritapossible/Tare
built with @openservai SERV Reasoning
Kite track. @GoKiteAI
Will retweeted
apparently, not updating your iOS for too long can leave your device exposed to security vulnerabilities.
oh well, i’m still holding out because that new battery icon sucks so bad fm.
where those hackers at? come find me cos i’m clearly not updating anytime soon 🤷🏻♀️
Will retweeted
Bro is really having a good time on @1win
Watching this makes you wish you were the one getting those wins.
Beautiful session to watch. Hopefully, my turn to enjoy some luck comes soon too.
Will retweeted
wondering how well the product details hold up when one image gets turned into several different ad creatives @getryze
ChatGPT ads are now 4x cheaper than Google or Meta Ads
It's early TikTok ads all over again, first in prints money 👇
Launch them from one prompt with Ryze AI:
-> Type "Launch ChatGPT ads, $50/day"
-> Ryze AI writes the ads
-> Builds the campaign
-> Picks the audience
-> Waits for your approval
-> Tracks every sale
-> Shows CPA next to Google and Meta
-> Finds wasted ad spend
We use our AI on dozens of clients we run ads for
Try it free: ryze.ai/chatgpt
Will retweeted
being able to spot underperforming ads just by asking a question instead of manually going through reports is pretty useful @getryze
ChatGPT ads are now 4x cheaper than Google or Meta Ads
It's early TikTok ads all over again, first in prints money 👇
Launch them from one prompt with Ryze AI:
-> Type "Launch ChatGPT ads, $50/day"
-> Ryze AI writes the ads
-> Builds the campaign
-> Picks the audience
-> Waits for your approval
-> Tracks every sale
-> Shows CPA next to Google and Meta
-> Finds wasted ad spend
We use our AI on dozens of clients we run ads for
Try it free: ryze.ai/chatgpt
Will retweeted
Sometimes the news moves faster than your ability to find the right market.
That's an interesting gap @swaydotfun is looking to close.
Will retweeted
2AM & my mutual @Juli_an_1 is on space 🤨
Tomorrow she'll tell me she's not a witch 😒
Case adjourned!!
Crypto doesn't just move on numbers. @swaydotfun shows that Narratives, conversations and collective attention play a huge part too.
Will retweeted
Bing Bong! 🔔 London is calling, and Phantom is bringing the Solana energy to Breakpoint.
@joinbingbong After Dark is taking over Soho during solana London Breakpoint. Same energy, a new night, and another chance to bring everyone together.
Something tells me this is going to be one for the books🇬🇧
Will retweeted
His excellency @officialABAT stay aso rock😡 sir
Your coming to my state has stressed me out since yesterday. Unnecessary traffic jam😡
I literally treked home yesterday, today I'm stuck in 2hrs plus traffic & I'm not even close to my office😡
With all due respect stay Abj
There's an interesting connection between narratives, market attention and trading. @swaydotfun is worth a look
Will retweeted
I really wanna know, can't we just stay for 30 days straight without sleeping?¿
Must we sleep sef 🤧
Some of the most interesting market ideas start as conversations on the timeline. What happens when those ideas become markets?
@swaydotfun is exploring exactly that intersection
Will retweeted
One thing I find interesting about @bootrfun is how it's giving AI a real role in the Web3 space.
With $BOOTR , launching a token can mean more than just creating a coin and hoping people notice it.
Every launched token can have its own AI agent with an X account, an on-chain identity and a treasury.
These agents can share updates, interact with the community, monitor market activity and earn fees to support their operations.
I like this approach because keeping a project active and engaging with holders takes time and effort. Bootr is exploring how AI can help with that.
It's still early, but I'm interested to see how far this idea goes.
AI and Web3 are an interesting combination. $BOOTR is definitely one to watch.
Will retweeted
BUSINESSES & CASHING OUT
There are five fronts in Cattown... The café, salon, sushi restaurant, jazz club and casino –– each with their own rarity and fee distribution tiers tied to our NFTs and the holders who own them.
Use them to launder in-game earnings into tokens.
Will retweeted
🚨 THE COUNTDOWN IS ON! $SQUEEZE IS READY TO MAKE SOME NOISE! 🐙🔥
We’re going LIVE on Pump.fun at 17:00–18:00 UTC!
No fake promises, no empty hype. Just a growing, active community and a project ready to make its move. 💪
The community is alive, the energy is building, and the next chapter is about to begin!
👀 Early to discover, early to position.
💎 Don’t just watch, be part of the journey.
This is a BUY NOW OR CRY LATER kind of opportunity. 🔥
📢 CA drops immediately after launch in our Telegram, so make sure you’re inside and ready!
🐙 TG: t.me/SqueezeOctoSquad
🐦 Follow our X:
Get your bags ready. Let’s meet at launch! 🚀 teapawt.org/SqueezeOcto
Will retweeted
me discovering there's a whole market attached to the random topic everyone has been arguing about all day @swaydotfun
Will retweeted
You shouldn't have to expose everything just to prove something is true.
That's the problem I find interesting about @primus_labs
Most digital systems make verification feel like an exchange: you want someone to trust your information, so you hand over more data than they actually need.
Primus is exploring a different approach through two technologies:
→ zkTLS: Verifies information from Web2 sources without exposing the sensitive data behind it.
→ zkFHE: Allows computations on encrypted data without revealing the underlying information.
Think about the practical difference.
An institution could prove financial eligibility without publishing its entire portfolio. An AI agent could process trusted information without needing unrestricted access to private user data.
That's where these technologies become more than technical concepts. They could change how trust works across DeFi, identity and AI.
What I find compelling is that Primus isn't treating privacy and verification as opposing goals.
The aim is to make information useful and verifiable without making everything public.
That's an important problem to solve as more of our lives move onchain.
Explore Primus and complete $PRIM opportunity XP tasks: s.kaito.ai/AUnKYVl
Will retweeted
London is getting ready for a different kind of night 👀
Bing Bong After Dark, powered by @phantom, is bringing the energy to Soho during Breakpoint London.
The IRL countdown is officially on. Keep an eye on @joinbingbong because this one's shaping up to be a proper experience.
London, don't miss out ☺️
#BingBong #Phantom #Solana #London #Breakpoint
Will retweeted
In trading, knowing your exposure is just as important as knowing your entry.
Smart decisions start with clear visibility. Stay informed, stay in control, and trade smarter with Bitunix Ultra
Jobless Claims can leave multiple TradFi orders moving at once.
With Bitunix Ultra, see filled vs. total order size clearly by quantity or position size, so you know exactly where your exposure stands.
👉 Explore TradFi with Bitunix Ultra: bitunix.com/p/9z9f
♤ What if losing a bet could still get you closer to ownership? ♤
Sounds unusual, right?
That's the interesting idea behind Stake Market.
@stakemarketonx is a prediction market where you back YES or NO on real-world events Built on Robinhood Chain .
So here's the twist:
● Make the right prediction? You win a share of the pool.
● Get it wrong? Your verified net losses earn you Points toward $STAKE purchases by the treasury.
Every Point represents $1 in treasury budget for buying $STAKE, with tokens becoming claimable after allocation.
It's not a refund for losing, but a different way to turn participation into potential ownership.
Win the bet. Lose into ownership.
Which makes this definitely an interesting approach to prediction market worth looking into.
DYOR .
Will retweeted
What is @primus_labs bringing to the Web3 ecosystem?
Primus is building advanced privacy solutions for onchain finance and AI agents through zkTLS and Fully Homomorphic Encryption (FHE).
The project has secured $6.5M in funding, with backing from YZi Labs, Samsung Next, Dispersion Capital, and other investors.
Want to know more about what Primus is building?
Explore here: s.kaito.ai/VzYCenp
Connect your wallet and install the official Primus extension.
Complete available tasks to build your Reputation Score.
Explore Primus Vaults to shield USDT and potentially earn yield and points.
Check the current terms and risks before depositing.
Can you prove something without revealing everything?
That's the challenge Primus is working to solve.
Some solutions keep data completely hidden, while others make everything visible. But both approaches have limitations when handling valuable information and important decisions.
The real breakthrough is being able to verify information without revealing the actual data, and process sensitive information without having to decrypt it.
This is exactly what @primus_labs is working on.
The goal is to make privacy and usability work together.
Keep your data protected.
Prove what needs to be proven.
Use information without exposing what should remain private.
That's the future Primus is working to create.
Explore Link: s.kaito.ai/VzYCenp

























