Pinned Tweet
1/6 @Mastercard just launched Agent Pay for Machines. New infrastructure built for AI agents to transact with each other, autonomously, at machine speed.
@Nevermined_ai is part of it 🧵
We launched the Nevermined Catalog on Product Hunt today.
We tested ~2,000 pay-per-call AI services and kept less than 200 that actually work. 162 services, one MCP, your agent pays per call by card, USDC or credits.
Tell us what broke:
producthunt.com/products/nev…
There's been a lot of chatter the last few weeks in agentic commerce about @Stripe going after the @Visa and @Mastercard kingdoms. @dongossen has a different take...Stripe is moving to take over deposit accounts, not the card rails. Unpacked below.
Nevermined retweeted
Last week I reviewed @Nevermined_ai and flagged what was broken.
Then I did something more useful this week: I gave my AI agent the 1,279 credits I had left after it was fixed.
It spent 418 credits. About 42 cents. Across 74 paid calls. No card. No top-ups.
Some of the problems are already fixed. 🧵
I gave an AI agent 2,000 credits and 90 real API call on @Nevermined_ai. It spent 708 credits, about 71 cents.
Disclosure: Nevermined provided the credits for testing. 2,000 credits is about $2 at their own pack price ($10 for 10,000). No card, no wallet, no top-ups. Everything below ran the way an agent would call it, through their REST API and MCP server.
What I tested
Easy: IP lookups, a weather check, geocoding.
Typical: a research agent across three search engines plus a scraper, a due diligence check on a real website, an SEC filing search, a DeepL translation, AI images, and auto-pick, including asks it can't do.
Hard: bad requests, wrong paths, oversized requests, the same request sent twice, an LLM call, a fake API key, a full MCP run, a one-prompt "US Creator Weekend Brief", a 40-call reliability loop, and a full ledger audit.
What it does well
➢ Billing is exact. Every charge in the ledger added up to my balance drop, to the credit. Every quote matched the charge.
➢ Reliability was near perfect. 88 of 89 valid paid calls came back. The one failure was refunded automatically and the retry worked. The reliability loop went 40 of 40.
➢ Failed calls are mostly fair. A bad request was refunded. Wrong paths, missing fields and oversized requests were refused before any charge.
➢ Duplicate protection works. Same request twice, one charge.
➢ The one-prompt test is the real win. Two search engines, a scraper, an LLM, DeepL, an image and file hosting: 8 calls, 93 credits, 79 seconds, zero vendor signups. It found 3 real upcoming US creator events, with a Spanish hook and a hero image.
➢ It's built for agents. The MCP server tells the agent its balance, the cost and when credits expire, in plain sentences.
➢ It says no honestly. "Book a hotel room" came back as no match, charged nothing.
What it does badly
➢ Slow. Every paid call takes about 4 seconds. The same IP lookup straight from the vendor took 0.2 seconds.
➢ Headline prices are off. DeepL's card says half a cent. I paid 5 cents, 10x the headline.
➢ The markup is hidden. On average I paid 1.68x what vendors received. The LLM call was about 24x.
➢ You pay for junk if the vendor says 200. A paywalled scrape, two useless SEC searches and a geocode that put a well known neighborhood in the wrong country were all charged. The brief's LLM also got one event date wrong.
➢ "No stablecoins" covers part of the shelf. Credits pay 135 of 160 services and 61% of endpoints.
➢ No spending cap on credits. A looping agent could drain the whole balance.
➢ Dashboard rough edges. It says "No purchases yet" while Purchases lists every call, and "Total spent" shows a blank dash.
What 1,000 credits (about $1) buys at measured prices
➢ 500 IP lookups
➢ 333 page scrapes
➢ 200 AI images
➢ 125 short GPT-4o-mini answers
➢ 100 Exa searches
➢ 20 DeepL translations
➢ 12 website screenshots
➢ About 10 full US creator briefs
Scores (my judgment)
Onboarding 8
Discovery 8
Agent experience 8
Billing accuracy 10
Reliability 9
Failed-call fairness 7
Price transparency 5
Latency 4
Coverage on free credits 6
Budget controls 4
Docs accuracy 6
Verdict: 7/10
The plumbing is the best I've tested. Exact money, almost nothing broke, and one key replaced 13 vendor signups. It loses points on speed, headline prices that don't match what you pay, and charging for useless "successful" answers.
➢ Use it if you're building an agent that needs many data sources without signing up for each one.
➢ Skip it if you need speed, heavy use of one vendor, or a hard spending cap without a crypto wallet.
What I'd fix first
Show the real per-call price on every card
Add spending caps for credits, not just USDC
Cut the 4 second overhead per call
Pass vendor error messages back to the agent
Fix catalog paths the router refuses
Let agents flag junk 200 answers for a refund
Would you let your agent spend money on its own? Tell me below 👇
Agents now pay for services on their own. When a purchase is questioned, who can prove what happened?
The AEP consortium, announced by A-Comm, was set up to answer that: one record sellers, payment providers and agent platforms can all check.
Nevermined is a founding member.
6
24
463
One of our engineers just gave an agent a $40 cap and it bought its own website: domain, hosting, preview image, voice and music!
It made 41 payments across 4 vendors several payment protocols. The Router picked the rail for each payment and every line on the receipt links to its transaction.
I gave an AI agent a budget of $40 and one job: put its own website online, paying for whatever it needed. It bought a domain, hosting, its own preview image, then the voice and music for this video. 41 payments. 4 vendors. Zero human clicks. youtu.be/Wg-7go30WRs
Nevermined retweeted
I gave an AI agent a budget of $40 and one job: put its own website online, paying for whatever it needed. It bought a domain, hosting, its own preview image, then the voice and music for this video. 41 payments. 4 vendors. Zero human clicks. youtu.be/Wg-7go30WRs
So I wanted a weekly brief on agentic commerce without opening ten tabs every Monday. I built one as an @n8n_io workflow and gave it a @Nevermined_ai budget to pay for its own research.
Demo: youtu.be/DEHwDbTcw-Y
1/ Your agent paid for the image. Until now, it could not pick it up.
Async AI services take the payment and hand back a job id. Come back for the result and they answer with a 402 that asks for no money, only a sign-in as the payer, so a leaked job id can't leak the image.
3/ Nevermined now signs it for your agent, only as the account that paid that same listing, within an hour of paying. It checks every field first, builds the message itself (the service's own free text is dropped), and the check comes back paid: false.
1
49
4/ No key to hold, no signature to write. One paid call, then as many checks on the job as it needs. Live on 8 of 161 catalog services.
The walk-through, with a real challenge decoded field by field: nevermined.ai/blog/your-agen…
29
Nevermined retweeted
Today is probably a good time to check in on things with @privy_io, which became part of @stripe just over a year ago.
Our acquisition was predicated on three factors: stablecoin growth, self-custody as a key enabler of broader global coverage, and rapidly increasing demand for programmable financial infrastructure. As we said at the time, "money has to reside somewhere, and Privy builds the world's best programmable vaults."
Since then, the Privy team has executed exceptionally well: assets in Privy wallets have grown ~25x, monthly active developers are up ~3x, revenue is up ~9x, and Privy now facilitates nearly $25B of monthly transaction volume. @tryramp, @Deel, @Klarna, @felixpago, @HyperliquidX, and thousands of other businesses are building embedded banking functionality on Privy.
At Stripe, we're deeply integrating Privy into Treasury, @Link, and Connect. In each case, it's accelerating our global expansion, and enabling us to build better functionality faster. (Coming soon: @Link balances.)
We're also seeing that Privy's wallet infrastructure is the ideal substrate for agents to transact and earn -- live today with @merit_systems, @orthogonal_sh, @Nevermined_ai, @prlnet, and more. Developers prefer programmable financial infrastructure, but agents need it. We expect that the agentic economy will continue to grow extremely quickly.
The Privy team has maintained remarkable product velocity, with major launches this year including deposit addresses, earning on Privy balances, digital asset accounts, custodial wallets, and Stripe's own stablecoin cards.
It's also been neat to see how Privy has been accelerated by Stripe, with better onramps, @stablecoin payment rails and licenses, Stripe's fiat capabilities and issuing stack, and integration with @Tempo.
Together, we're building one API for holding stablecoin balances and earning yield, moving money in and out of fiat, and orchestrating spend with cards. We still have a lot of work to do, not least in integrating everything across Privy, Bridge, and the rest of Stripe -- there's been so much feature and functionality expansion that we haven't yet had much time to reconcile everything. Overall, though, I'm very happy with how things are going, and excited by where these capabilities are headed.
If you're interested in working on the future of stablecoins, agentic payments, and crypto's deployment in the real world, you should get in touch with @sternhenri (now leading all of @StripeCrypto) or others on the team.
Your API works. Can an agent use it? 🤔
In the Nevermined AI Services Catalog we already check that a service is alive and can take a payment. That doesn’t tell you whether an agent, with no human next to it, can actually use it. 🧵
Readiness is not quality, popularity, or traction. A service shipped yesterday with clear docs can score higher than a busy one with none. It only flags the gaps that stop a useful service from being found, called, or paid.
1
2
75
It’s advisory and never blocks a listing. See it on each service’s page, or as readinessScore in the Nevermined Catalog feed.
Agents are the new buyers. Build for them 👇
nevermined.ai/blog/whats-new…
2
31
Intelligence isn’t the bottleneck anymore. Access is.
We gave an agent a budget and the Nevermined Catalog. It found, paid for and stitched 22 services into a live app for $1.82.
No keys, no signups. Read our blog post to see what happened when the agent couldn’t pay👇
nevermined.ai/blog/what-one-…
Nevermined retweeted
Replying to @Nevermined_ai
Paying for use makes software outcomes more tangible for everyone
Nevermined retweeted
AI agents can buy the capabilities they need while working.
@autonolas, @Nevermined_ai and @PayAINetwork show three useful approaches to that market:
1/ @autonolas: Hire another agent
Mech Marketplace lets developers offer specialized agents as paid services.
An agent needing a forecast can send a task to a Mech and pay for its response. The service provider earns from being hired.
The mechanism: turn an agent’s skill into something other agents can purchase.
2/ @Nevermined_ai: Put spending under a budget
Nevermined combines service discovery, usage-based charging and spending controls.
Its Catalog lets agents find paid services and call them within a human-set budget, with limits and a payment ledger.
In its October 7 team demo, an agent used 22 services for $1.82. That excludes the model’s own usage cost.
The mechanism: make buying tools part of the workflow, with spending boundaries.
3/ @PayAINetwork: Settle payments for paid requests
PayAI provides an x402 facilitator for stablecoin payments.
A service returns payment terms through HTTP 402. The buyer signs a payment, and the facilitator verifies and settles it on a supported network.
The mechanism: give APIs and agent services a payment path for individual requests.
These products overlap, but each highlights a different problem: finding skills, controlling spending and moving money.
The harder question comes after payment: did the purchased service produce something useful? A payment receipt alone cannot answer that.
This is very cool. @Muse was given a money "harness" so that it can't go off the rails.
Trust is a bottleneck to giving agent's money. Can I trust Muse or @Meta to ensure the agent doesn't go over budget? @dongossen just put spend limits on Muse that HE controls. Pretty rad!
I just put guardrails on this biatch!
@Nevermined_ai gives me payment controls that only I control (NOT @Meta or @Muse), so I can give my Muse money, but it can't go spend willy nilly.
I asked my Muse why this is important. Here's its response:

















