Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
SKALE Developer Showcase: Episode 2
Recorded: March 12, 2026 Duration: 0:55:11Space Recording
Full Transcription
100100100100.
For those of you that don't know, that's high in binary.
Welcome to Scale Developer Showcase.
Now, am I telling the truth?
Did I actually learn that, Manel?
Or did I lie?
I think you already did.
I did. I did. I'm really proud. I did.
How you doing, man?
All good. Super excited for today.
Incredible guests.
Yeah, it's going to be a fun one.
It's going to be a fun one.
For those of you who don't know,
this is the Scale Developer Showcase
where we bring forth uh partners and projects building on the scale ecosystem
to show you all and us as well what they're building how they're building it and why you
should be excited now our first guest is coming over from primer mr, Mr. Joe. How are you doing, my friend?
Hi, guys. I'm good. Pleased to be here. How are you two?
Oh, great. So pretty excited.
Pretty good.
Yeah. My back's hurting, but yeah, not too bad. I need a new mattress, man. Well, Joe,
why don't you give us a little introduction about yourself, Primer Systems, and what you're
going to be showing us today.
Sure.
So a bit of background on me.
I've been in crypto since 2013.
Seen a lot of things come and go.
I've done a lot of things.
I worked for a while as a CIO at a hedge fund, a crypto hedge fund.
And through that, I started helping our portfolio companies in various ways. That became more and more full time and then more than full time. And I kind of
gradually got into doing some dev work with them as well. And long story short, that's led me
through to discovering X402 and some other things last summer and getting on to the trial now so Primer was launched in December
and it made use of certain technologies that I had seen as coming up and becoming important and
looked really interesting to me last summer X402 being quite new I wanted to learn about it say in
August September and as usual the best thing to do to learn about something is to try and build on it
and figure it all out.
So we built a facilitator.
And my first thought with that back then was, well, it's a bit limited, isn't it?
Because it's pretty much only USDC or EIP 309 compliant tokens.
As it happens, I no longer see that as a limitation.
I think it's probably fine, but we decided to build a way to
accept any ERC20 tokens through a kind of router contract that we now use called Prism.
We did a few other things anyway, and then launched Prima officially in December. And
we'll go into more detail about the exact product momentarily, I think.
exact product momentary, I think.
Yeah, yeah, to be honest, I think your product is, first of all, is one of those that I still
need to try myself because I'm really excited. I really want to go very deep on your movie
ball and all that sort of stuff because, yeah, it's different. And I think when there's something
different out there, I think it's always, it's good in the sense that you just didn't build...
You started by building a facilitator.
There's a lot of facilitators out there, right?
But now you just built a different thing.
And I don't know if you're going to showcase Multiclaw.
But, yeah, I really like your product.
Sure.
Well, there's two of our main flagship products
of Primer Pay and MultiClaw.
I think we could probably got time
for a quick demo of each of them.
Absolutely.
They represent quite different sides of X402.
So let's look at Primer Pay anyway, first of all,
if I can share my screen.
Here we go.
You got that up?
Yes.
When we started thinking about what we could build,
everyone's talking about X402 for Argentix,
which is all well and good, and we're a big believer in that.
But we thought there was a bit of a gap in that no one was really addressing how x402 can be used by people as well
and if you stop and think for a moment how much how many different places we spend money online
it's probably best we don't think about it really there's all kinds of subscriptions and streaming
and everything as well as normal shopping and i think X402 has huge potential there actually.
After all these years in crypto, there is no really good fast way to pay for things with it.
There are services you can do, but it's cumbersome. It's very difficult for the merchants in particular
to set up. Usually it involves some kind of token conversion and then they get fiat in their bank
accounts and there's quite high fees and so on, several percent, you know, on a pile with cards and things.
So we've built a browser extension for Chrome. It's on Chrome Web Store now.
I've got it installed already, but you can find it on Chrome Web Store.
I'll show you a little overview of it.
Normally, you would at this point set up your wallet in the usual way that you're familiar with MetaMask and so on.
Here's our balance on scale base. We've got five cents, which is sufficient for today.
The settings here, we've got a per payment limit, $1 at the moment. I'll just up this daily limit to $1.
We set it to auto pay or you can have manual
confirmation at different thresholds. Got your networks, domain, block list and things
like that. You can have an explorer of that if you like.
So to test this out, because the internet is not yet full of exploratory paywalls that
humans can interact with, but we think it will be sooner or later. I've x402 paywalls that humans can interact with but we think it will be
sooner or later i've got a test paywall set up here on our website so um oh and if anyone can
play with this if we've got it uh you can just put in your own address basically it'll loop your
money back to you so you can you can have a play with this i'm going to disable primer pay first
okay and then when we try to access his paid contact we get a 402. That's a dead end. You can't go any further. If we switch PrimerPay back on,
with any luck, it will intercept the 402, process it, and there we get to the paid content and we should see this transaction on your block
explorer there it is um i'll give you an example of of uh what would happen if we if we fail
somewhere so let's set the daily limit to one cent we've just spent one cent so this should
not go through anymore there we get a it It's blocked. You've exceeded your daily net.
That's a simple example of what PrimerPay does.
The main thing we're building in this part of our ecosystem next is the merchant side,
which I think is sorely lacking because you can set up an X402 Paywall as a merchant,
but it's not very straightforward.
There's no wizard really to do it.
We want simple plugins, WordPress kind of things.
That's the kind of stuff we're building for this.
And then that closes the circle basically
on human interacting web browsing.
So that's one side of things we're really interested in.
Do you wanna talk a bit about Primer Pay now
or should we move on to Multiclaw?
Any questions on that?
It's very, very interesting.
And like I said, you built different projects
that I haven't seen it yet.
That's why I like it so much.
And I'm really excited to see where you end up,
and are you going to approach getting projects using you and that sort of stuff.
So yeah, I would say let's move to Multiclaw.
Yeah, I'll then make questions.
Sure.
Okay, so Multiclaw is completely different.
The origins of this, so we took part in your X402 hackathon a few weeks ago, which was fantastic.
This was our submission.
We called it an agent manifold.
And after the hackathon, we developed it into Multichlor as the full production version.
We're close to release of V2.
We're still going to use v1 for this demo
because we're still ironing out a few creases. You can find it on GitHub. It runs on your desktop or
USB. It is an app, looks like this. I've already set up a lot of things.
Of course, I've got a wallet set up.
You can import keys,
low keys, derive from seeds and things.
First of all, we have spend policies,
the crucial part.
At this point, I should add,
why did we come up with this in the first place?
That earlier this year,
and I'm sure there'll be more examples of this there were
lots of quite public cases on x and so on of people losing money to um misbehaving agents
and I'm not surprised frankly you know we we use agents all the time for all kinds of things and
I don't I think at this point they can't be trusted essentially for these kind of things they
they are consumer products right they want to please you they make money if you keep using them
and renew your subscription and uh they are yeah they can be deceptive and sycophantic and you
can't really if you say do not share my key do not spend more than this it doesn't really mean
anything we think you need hard separation at for now, between your keys and the agent.
So multi-claw is designed to sit in the middle of your keys
and your agents.
Low friction, but complete separation.
So we'll set up a spend policy.
We can just call it number two.
Choose your network.
You've got a daily limit, per request limit,
we'll set auto-approve. And you've got a domain limit per request limit all sets auto-approve and you've got a domain
allow and block list as well then we register an agent now an agent can be literally anything this
will work with any agent that you can communicate through by text which is any agent okay we're
going to use standard clawed instance today okay you register your agent. There's two different auth modes. We'll use a simple
bearer token today to speed things up. And you get this code snippet that you're going
to pass on to the agent. Click to register the agent. We're actually going to use a V2
version of that today for better performance. We're going to go to a new Claude instance here and just paste that snippet that I gave it.
Now, you might think that's not very many, not much instructions for the agent. What we're
actually running here is a local server with a full skill file and all the information the agent
needs to get going. And he has his security stuff, his token, and things like that here.
So while the agent gets started on the instructions,
we have to commission the agent.
So an agent starts like in an uncommissioned state.
We commission it by attaching a spend policy
and a signing address.
At this point, we can generate an AP2 intent mandate as well,
which was a key part of our hackathon submission.
If we go to the,
we'll come back to the mandates in just a moment.
So Claude has figured this all out
and we're going to give him a paywall now.
So I've got a, this is a different test paywall builder. We've set one up for one cent and I've got a URL here.
I'm just going to give him the URL.
And he should figure out based on those instructions what to do with it.
what to do with it.
While we're waiting, I'll talk a bit about V2
if you like the main difference, main additions.
We've added a console feature here
so you can interact kind of CLI
and there will be true CLI and headless modes as well.
So I really like this because like in the end of the day, Multicall is like a control center where you can control and make sure like you control the agents, what they can
do, what they can spend, right?
Yeah, exactly.
You can connect hundreds of wallets, hundreds of agents, hundreds of policies in any combinations that you like, essentially.
At the moment, it's X402 only.
We're working on ways to expand that to non-X402 payments as well.
Manual transfers, maybe DeFi things and everything like that.
That's obviously a little bit more complex.
The beauty of X402 is that it is so simple in a way,
but Multiplore will be able to handle
all manner of things like that.
Here's the new transaction come through.
So the agent has reported it signed.
He's gonna to go back
to the paywall. Now I should add, we're using a stock clawed code, which is not really set
up for this thing. Whatever agents people are running in particular, this can be made much more streamlined.
We've got it set up to be quite verbose
and quite interactive,
which wouldn't be necessary in most situations.
He's going to report the transaction as settled.
Here it is.
We can verify that.
And I believe if we go to the Block Explorer,
we'll see that one come up as well.
There it is.
That's, yeah.
Very nice, very nice.
One quick question that I didn't catch up.
How you set up the agent is like you passed the,
what exactly how you set up the agent?
So when you register an agent, you just get get this code snippet i got it okay and
you just pass that to the agent copy and paste um so what i said to claude at the start here
this is literally all i said to him so this is a completely fresh cord with no no prior knowledge
of what what we're doing at all you just paste it says, I've set up a local service called Multiclaw
that I want to interact with, a few other things.
And then there's this local URL
which contains the full instructions,
customized full instructions.
So the agent reads that,
he's got his credentials and things here,
and that's all you need to do.
You don't need to explain X402
to explain x402 is or anything like that it's all contained in this um and you get a unique
or anything like that, it's all contained in this.
uh config that you pass on are those um what i'm thinking now is like when creating an agent is it
um can i configure like can i adjust that agent context and, like, what the agent will look like for
me or is it, like, standardized?
Well, so, Multiclaw doesn't contain the agents itself, okay?
It's completely standalone.
You can use this with any agent whatsoever.
So, whatever kind of agent you're running, whether I'm using just a stock claw code for the demo,
or if you've got an open claw or anything else, any agent that has internet access and can
understand text will set up the controls that you have. So an agent, as far as multi-claw is
concerned, is basically a set of configs that you attach to whatever agent you want and
then corresponding spend policies and so on.
Yeah, pretty amazing to be honest. I'll for sure work on this and then I'll talk.
Do you have any future plans for instance on the, I don't know if you worked with already,
like with 8004, like Scale recently just launched
8004, the official contract for registry
and reputation of agents.
I'm not sure if you have any plans on that
or if it's something you're looking to it.
When we launched it was um definitely on the cards to
to be honest we've become a lot more excited about x402 uh which i think is the more fertile
building ground at the moment we do have a few 804 projects um that we've put on hold for a little
while uh we really want to get multi-claw and primer pay and it's kind of accessory things
up to scratch first of all but there's a few 8004
things in the works and of course they go hand in hand quite quite naturally. Multiclaw we have a
way we've got AP2 mandates and things which is sort of a reputation thing if you like but
8004 connection with your agents would kind of work. This is for the user, the person who owns their agents to use.
So if you're running an agent, you probably don't need to see it's 804, right?
So that's kind of more between your agent and the merchant in a way.
In terms of other future plans, like you talked about Multi-Clor and Primer Pay,
we've been thinking a lot about Multi--chain X402, which I know some
of your existing partners are involved in, and how privacy can be used as well.
Both of those kind of go hand in hand. We have some interesting larger plans for those.
That's very interesting to say that we, I don't know if you've read about bytes and the encrypted transactions.
I'm not sure if I can talk about it.
I'm not sure if it's the confidential token that we are launching.
Yeah, you can talk about it.
I wasn't sure.
So yeah, I will launch ERC 20 confidential token that uses byte so
yeah that's something I think might interested you but we can check we can
chat up that about that later yeah it's funny we had a look into it we it was
one your tracks for the hackathon that we like the APC track of course in the
end but yeah I think butte's really, really interesting actually.
Something we'll definitely be trying to include.
That's great.
That's all on my side for Cork.
No, it all looks really, really interesting.
I guess like, let's play a prediction game here.
You know, we're talking about X402 and its potential, you know, use cases in the future.
You know, how long estimate do you see
until X402 is going to really start to take off?
Perhaps it already has.
We're kind of in it.
These things all come and go in
in waves a little bit right um but i think this year is going to see multiple waves of of x402
interest it a lot of it depends on the wider market and people's sentiment right how much they
they're excited about these things depends a lot on on prices and so on but um what when x402 is
is what uh nine months old or something at the moment.
It was like last summer.
It kind of came out.
May, March, May.
Maybe coming up to a year, which is incredible, really.
And it's really taken off.
It's got its own categories on CoinGecko and so on.
And loads of people are building on it.
I think it's a really attractive area.
It uses the unused HTTP code from the 90s. I think it's a really attractive area. It has, you know, it uses these,
the unused HTTP code from the 90s
and it's such a pure, simple technology
that I think it really appeals to a lot of different builders
and it's almost inevitable.
I mean, we've got massive backing,
got Cloudflare, AWS talking about it now.
It's, yeah, I think it's inevitable.
Yeah, my opinion, I think it's just a matter of, like, you already said, like, we have
the right companies backing it, and then if we get, like, that product that just unlocks,
like, real usage, I was right now thinking, like, imagine that ChatGPT would, like, would
have come out with, instead of, like, being that Chatbot website like would have come out with instead of like being
that chat bot website would be like a CLI like no one would use it and apart from the devs right
I think it's if you get that user experience right using Xpro2 with agents and just have them
pay for resources that benefits the end user, I think, yeah, we'll just...
Yeah, that's an important general consideration, you know, as much as we,
as devs, we like building things that might be CLI and easy to use for certain situations,
having a nice friendly graphical interface of some sort is going to be really important in
getting wider appeal, which is something we've tried to do to our approaches yep yep absolutely yeah hell yeah um well joe
really really appreciate you taking the time to run us through primer and everything you guys are
doing anything further before we let you go no i'm very happy. It's great to be here. I want to thank Scale for your support before, during, and after the hackathon.
Having worked across lots of different chains,
dev support is generally ranging between OK to actively hostile,
and Scale has been really fantastic.
So it's great that what you're focusing on being so similar to what we're working on
and the support you give to your devs has been really good.
Looking forward to the rest of the show and what your other guests are doing.
Appreciate it.
Appreciate it.
And if you want the hostility back,
I can talk to Sawyer and get him and just get him grumpy.
Yeah.
I want to have him.
Awesome.
Thanks so much,
Joe.
And yeah,
we're excited to keep on seeing what Prime is doing.
Excellent. Pleasure, guys. See you.
Have a good one, man.
Bye.
Oh, Manel's got the coughs.
Manel is ill.
Sorry.
That was cool. I like that.
I told you.
It's a different good.
Yeah. Yeah.
There's lots to...
Definitely lots of use cases,
which I like.
And I think there's also
lots of potential integration pieces
with other projects,
other partners,
and things like that.
So, yeah.
I agree.
And I think that these kind of things
are important,
like new things.
Even if it's not
the thing that's going to work
like just generating ideas
building
it's not actually that that will work
it's the thing on the side
that maybe you need to just beat it a bit
it's very important
for us for the product
for the whole space
100%
so Manel while we wait for our
our next guest to show up um why don't we actually touch a little bit more on conditional transactions
right i think that could be a that could be an interesting topic just to just to quickly go over
obviously it's something that we haven't released to the
public yet something we have spoken we have spoken about um we had the hackathon on it but
why don't you give the give the peeps a little bit of an insight into conditional transactions
absolutely so it's a super interesting thing that's engineering,
the scale engineer team built.
So basically, I don't know if you already use it,
but we have ByteV one, right?
So basically where it's possible to encrypt transactions.
So when you are building a transaction,
you can encrypt it on the client side
and sending that encrypted transaction to the chain and only
when it's on consensus, it's decrypted and then it gets on the chain decrypted.
And conditional transaction, what it does is it decrypts not on the consensus, but after a certain specific condition is met.
So you can trigger basically that decryption on the transaction that's in ByteV2.
And so let's talk use cases here, right?
What does this potential new tech really open up?
I mean, we still probably, I mean, I'll talk about use cases
and maybe I will just say things right on the surface
because there's so much potential.
But we've seen a lot of different use cases on the Akathon.
One other thing was like, imagine like a locker or vault. It really depends. But yeah, a vault that's something you want to invest in and you don an investment there we encrypt it and just after a
certain period of time it gets decrypt and you get access to that uh that vault for instance so
i mean that's different things to be honest yeah yeah i think some of the things that could also
be interesting is like and i don't know if this is impossible but but like, obviously contracts are, I guess you could argue contracts were kind
of a way to do conditional based things on the blockchain, right? I'm not talking about
transactions. I'm just saying things, you know, and, and, but the issue with contracts are,
they can be, if the contracts haven't been uh what's the word like
given up they can be rewritten right you can change contracts of tokens and things like that
yeah yeah yeah upgradable contracts while conditional transactions aren't going to be
upgradable are they that's that's a great question i would say no no no i would say no
but that's something to double check with oh no i don't think so to be honest i i don't think so
i don't think so either because then it's not really conditional anymore like you set up your
conditions prior to the prior to the transaction right and i think one of the things that interest
is interesting is it kind of opens up this obviously trust is something that is
uh still scarce in the world of web 3 but when you have something that is baked in that is
a hundred percent set in stone you are kind of pulling away that that sense of
lack of trust right for example you know what who would really benefit from something like this
is like a polymarket or something like that, right?
Where you have conditional based transactions.
I guess, is there a way this, this is like old scale school type stuff now,
where I just be like, so yeah, cat, is this possible?
And he's like, I don't know.
Like, could you do external conditions to a conditional transaction?
What do you mean external conditions?
Let's say...
Let's say I wanted to do...
Hypothetically, if the S&P were to drop five like five percent
this transaction will execute yes so I guess what it does is it opens up you know real world world use cases, right? It's not just on chain related things.
You can open up the possibilities.
Like I said, it's something that we probably,
it's like we've built around the,
basically the byte V1 and by p2 was built around encryption and
what we are able to do with the way the chain is uh set up right yeah um because um the way the
consensus works between the nodes and the way they sign uh each each block uh it was we are we were able to build that with all the maths and cryptography around, so
it's mathematical provable, decentralized, and so on and so on. So it's something that
you don't get in all the things that they say they encrypt out there. So yeah, what I meant by this is like,
we've seen, we've thought about certain situations
that you might want to use conditional transactions,
but probably, and we hope that,
really hope that our devs come here and like think,
okay, you can actually use it in a different way
that we didn't think about it
so i just uh start i think that's the good thing about hackathons we've seen a lot of different
use cases i've talked about amazing projects coming out of the hackathon yeah i've talked
about the vault thing but we have another one that basically if i'm not mistaken what it was
if I'm not mistaken, what it was,
is like it was a battle of agents making trades,
something like that.
And the trades, they could get encrypted
and just get decrypted at the end of like
a certain period of time.
So no one would be able to copy the trades uh right um
i think that's that's that's pretty cool to be honest yeah yeah yeah no agreed agreed uh yeah
we've seen like lots of you know stories coming out about know, balances and things like that being, you know, big,
big issues at the moment.
And I think the beauty of this is that we are able to do this.
We are only able to do this because we set up the chain in a specific way that other
chains are not able to do it here.
I mean, it's not able to that polygon. Ethereum is not able to do that polygon, I
don't know, all the other chains. Because at the end of the day, we have these leaderless
nodes that because the consensus was put in this specific way, we are able to sign the
transaction and to create these. So basically, it's each node as a share, a secret key that then signs the transaction
and creates this is able to encrypt the transaction.
Yeah.
Yeah.
No, incredible.
Incredible.
And, you know, we've been yapping, you and me.
Maybe it's time to let someone else yap.
Do you have it there?
Perfect.
Yeah.
Hello, fellow yapper.
Hello.
You're talking to me?
I am.
Hello, Julian.
How are you doing, man?
Good.
Good to see you guys.
How are you?
Incredible. Now that you are here.
Yeah, I was getting sick of fucker.
No, it's how most people.
That's not true.
My wife said the same thing about 10 minutes
after me waking up, so.
I love it.
So Julian, why don't you give us a little introduction
about yourself and Cred Protocol.
Welcome to the show.
Yeah, thank you for having me.
So, yeah, Julian Gay, founder and CEO of Cred Protocol, based in San Francisco.
Yeah, Cred is an on-chain analytics platform focused on counterparty credit risk. So we, you could think of us as
effectively a crypto credit bureau. So we monitor activity across 10 different blockchains,
30 different lending protocols. We developed the first predictive credit score,
predictive of liquidation default default, delinquency.
And so that our core mission is scaling trust and transparency in Web3.
And that's essentially what credit worthiness is, is like scalable trust.
And that has taken us into the direction of the, well, the activity that we're monitoring could be produced by an individual, a business, a bot or agent.
So that's taken us into the kind of agentic reputation space.
It's how do agents trust each other, trust they're going to receive the services that they think they're going to
when they interact with another agent or a tool.
And so the agentic kind of reputation goes hand in hand with agentic payments
so we're kind of moving from our kind of core kind of credit area which is kind of focused on
lending into this kind of nascent agentic reputation area and you know obviously scale is a great
platform to do that on because of the the one like
millisecond kind of penalty but also the the free transactions which are going to be needed when
agents are looking up the reputation for all the other agents they're interacting with their when
they're acting autonomously that's super cool uh to be honest, you were talking and I was thinking, yeah, like this reputation,
makes a little sense. Like for instance, I was getting, I was talking with my girlfriend about
buying a house and going to the bank and get a loan and basically the bank just like reviews all
of your history, reputation, whatever.
And probably it's kind of the same, right?
When you're handling with money,
you need to get the reputation and the credentials of the agents
or like the person, the wallet, whatever, that you are using it, right?
Yeah, for sure.
I think particularly with agentic payments
on the blockchain because of that guaranteed finality that there's no
there's no charge backs I know there's some stuff being worked on with escrows
and things like that but there's not there's not really charge backs like
there are in you know traffic so if you're making payment or requesting a
service from another agent you need to be pretty
sure you're going to receive what you think you are because otherwise you would have paid for it
already yeah particularly I'm talking about like x402 so yeah reputation becomes kind of increasingly
important in that context but not not just for the kind of trusting you're going to get what you think you're going to get
what the agent's going to get what it thinks it's going to get but also kind of differentiating
between different agents I mean if you've got your different agents different tools maybe like
translating a pdf for example there could be different levels of quality in you know in that
translation and at different price points. And they're all legitimate.
It could be a premium service, it could be a lower end,
but it does what it needs to do.
You need to be able to differentiate that,
or the agent needs to be able to differentiate that
to make the appropriate choice when it's, you know,
trying to get the outcome that it's focused on.
Absolutely.
Yeah, I'm super excited to get it uh to to check your demo uh
are you ready to for that you are so the the agentic piece that's super nascent the um the
the demo the demo i'm going to show is um we've integrated scale to accept x402 payments for our kind of credit services. So that's what we
have now, we've had for a few years that we're extending with these agent-specific reputation
services which are basically based on ELC A004. So what I'm going to demo is I'll give a quick overview of the product just to
kind of you know orient you guys and then I'll show the scale kind of payments work. Cool let me There we go, okay.
Let's, so this is the kind of, you know, present,
our, how we present right now, which is around credit risk.
There's different aspects to that.
I'll, and essentially we're a suite
of APIs. So if we take a look at the APIs, not super compelling from a demo perspective,
but these are basically core credit bureau kind of API endpoints. You've got your credit scoring.
And so if I can check out Vitalik's credit score basically, you can see that was like
super quick request. That's actually, you know, we have a lot of like caching at various levels but what we're doing is going on chain across
eight different blockchains in real time pulling the activity for vitalik um and
running that through our um credit inference model to produce this this score so this is
credit inference model to produce this score.
So this is a live score,
literally if there's a transaction in the next block,
you're on the score again,
it could be, should be different.
And it's kind of a live score,
which is a way, it's a step,
it's way better than traditional credit scoring,
which is quite often delayed by weeks or months even.
What's the max value for the credit score?
So range is 300 to 1000, which kind of maps to,
if people are in the US, FICO is 300 to 850.
So it kind of maps to that range.
It's gonna be kind of typically
slightly higher and then we've got the credit categories of excellent very good good moderate
low which which again map to how people look at kind of consumer credit scores so that's kind of
intentional to kind of make you know credit a bit more familiar for people, you know,
working with, you know, traditional credit scores.
I didn't know about these credit scores.
I've heard about, yeah, okay, makes sense.
Yeah, I don't know where you guys are based,
but the US absolutely runs on credit scores
on the personal finance level.
You cannot do anything. You can't, cannot do anything you can't I mean you
can't get a credit card which means you can't get you can't rent a car maybe you can't you can't
rent an apartment in some cases you might not get a job so it's a credit scores in the US are
incredibly important you can't work in government if you have a low uh a low credit
score oh wow super in in the uk in the us ah i did not know that that's that's a new one
it's because you're you're open for bribery okay i didn't know i mean it makes sense
um so the thing that's behind that score is, okay, this is the credit report,
which basically, yeah, it's doing it right now. It's going to those 10 chains, because it's not
cash. You'll see, you'll get a kind of real realistic look on performance. And I use Vitalik as the as the, you know, kind of demo account, because that's
basically the account that has one of the longest histories on chain. So if we can do it for Vitalik,
we can do it every every address out there. And what does it show in this report? Okay,
yeah, so it's here. So literally, that was that's Vitalik's account.
Or you know, one of them. What it's showing is, you know, net net worth. Well, one age of the
account 10 years, the last activity is 13 hours ago. The so it's okay, it's got some exponentials here so net worth let's say very high um
down down down in the past 24 hours uh you know i think as we expect in the current markets
um and yeah not not not much borrowing and lending activity on on this account at least um
and lending activity on this account at least.
I think some other aspects,
I mean, we also look at identity associations.
So the reason we do this is,
the higher number of associations there,
the more confident we are that this is a stable wallet,
it's not gonna be, it's not been just spun up.
And one, it would obviously, you know,
not have existed for very long if it was just spun up.
But also we have kind of higher proof of personhood,
basically, so we do look at identity attestations,
we look at asset breakdowns,
percentiles across the whole spectrum of accounts.
So is this the top 90 whole spectrum of accounts so is this is this
really top you know top 90 90th percentile of activity and then we have
a breakdown of assets across the different chains so the report is is
kind of very detailed it's it's you know And that all drives into what results in a single number of credit
scores. Obviously, a kind of blunt instrument, but it's based on this report.
I would love to discover now that, sorry, sorry, I would just love to discover now that
that the other Vitalik was doing some transactions on scale,
Vitalik was doing some transactions on scale, but yeah, that's not the problem.
but yeah, no, that's not the problem.
Yeah, that'd be a good one to look into.
So yeah, I mean, just to quickly kind of buzz through this
and get through to the X402 demo,
I mean, so you can also kind of, you know,
do specialist endpoints, so this is just,
if you just want to like look at the attestations.
And part of the use case as well is kind of Sybil detection.
Part of the use case as well is kind of Sybil detection.
You know, how likely is it that this is a unique wallet with a genuine person behind it?
And so we look at basically, you know, the graph of transactions, you know, what has actually been spent on this wallet?
You know, so this kind of gives us a Sibyl score,
which is zero to 100.
So 100 would be highly likely as a Sibyl.
So the lower the better in this case.
Let me, I just mentioned the transaction graph.
I'll just give a quick kind of overview of that.
So this you can kind of select an address and you
can see all the other addresses that this account has interacted with and the nature of those
transactions so you know what what were the tokens that were transacted the the use that we put this to one kind of detecting um interactions with maybe bad actor
wallets um which may you know kind of taint the wallet you're actually looking at um also
um in certain clusters you can see other addresses that are close to this one so that means if it is
if they're at the center of
this graph it means they have very similar activity which means they are most likely or
there's a probability it's the same beneficial owner so so we can kind of detect whether actually
we're not necessarily seeing this the whole picture someone someone said this is my wallet, but maybe they have other wallets. Ideally,
ideally they disclose them. It's not ours that, you know, kind of encourages people to disclose,
it would be the protocols that build on top of us, which are lending protocols, identity protocols,
wallets. So, this is the graph aspect. I think this is useful in so many different levels.
Firstly, I thought about lending protocols, like you said, but even for chains, for them to give away grants, in the sense that, okay, a project wants a grant and they just share the wallet for the the chain runs like wants to run like sort of a QIC on the wallet and
It checks that okay that wallet is kind of sketchy through your
Cred your platform, right? I think yeah, it's very very useful because you can track and now they do a QIC
Yeah, I mean and to I mean just along those lines and and to Falco's point
I mean there's a couple
of scenarios there one is looking at the wallets of the multi-sig signers of say a dow or a project
you know how reputable are they you know have they do they have like long-standing activity or are
they just like you know fresh wallets with no activity, you would know nothing about them.
So there's that, the multi-sig kind of reputation.
And then I think, you know, the stuff you're talking about, Manuel, the kind of using,
using, you know, reputation credit worthiness as a filter for kind of airdrops and, you know,
are you, are projects dropping or rewarding like legitimate
users versus you know, farmers that just spun up a wallet. So to be, although it's it's it
is a use case we haven't we haven't had too many kind of you know, too many interactions
on the airdrop from I guess, maybe airdrops are a little bit out of fashion these days. Could you also, sorry, my brain, I think it's the, I've tried your application, but now
I'm getting more and more, would it be possible to like ever like a Chrome extension with
your, that would link to the user wallet?
I'm just thinking that like, so I work as a, on the DS team of scale, right?
So I test a lot of projects that I do working into scale.
And I got, I used burner wallets,
but I got scanned like two or three times, not scanned,
scanned in the sense that probably I was distracted.
I signed something and they just,
my wallet got drained, right?
Probably if I would have some sort of a Chrome extension
that would, when I was about to sign, it would just tell me, like, would have some sort of uh chrome extension that would when i was about to
sign and just tell me like would run uh some verification i would pay for for that verification
next for two payments for that verification on the water or water contract that i'm about to
sign right and would tell me okay uh just be careful these guys already drained some other
wallets or you know this kind of thing I
think it can be a really good use case I would pay for that to be honest instead of um getting
uh stolen for 10 USD USDC or SET I would pay just yeah one euro one USDC? I think that will be amazing. Yeah, awesome use case. We are integrated with a couple of
wallets, but it's not exactly that use case, it's slightly
different, which is, it's more around lenders, being able to
see who they are lending to, but it's a similar kind of scenario. They want to know,
you know, am I interacting with somebody kind of reputable? So we'd love to work with you
on that use case. Yeah, I think the other way around makes a lot of sense because
yeah, I mean, it's just like protecting yourself, like you protect against Mav attacks, right?
I think it's kind of the same.
We're not the same, but I think that the mindset is the same.
Yeah.
Hey, guys, I'm cautious.
We are at time.
Sorry.
I'm excited.
Tangent.
All right. Let me show X4 to payments on scale then.
Yes. Yes. You're fine? Yes. Okay. Sorry, Julian. All good. All good. Great. All right, let me show X42 payments on scale then, just as a wrap up.
Yes.
Yes.
You're fine?
Yes.
Okay.
Sorry, Julian.
All good.
All good.
Great conversation.
I'll just plug this.
Anybody can go and get their own score, score.credprotocol.com.
You connect your wallet, you'll see your score.
Only you see it, but it's a way of making those API requests a bit more friendly. So this is, let me share my screen,
so this is credprotocol.try. So you can try all those endpoints without having a credit account,
and this is because we're integrated with scale. So again, I've got, I'm going gonna look up Vitalik's credit score I'm gonna send
a request I receive a x402 message saying payment required I'm gonna count
my wallet yes you partially see this on screen oh I should fail live demo try that again all right so I'm connected so I'm gonna switch to
to scale it's asking me to switch network which I am I'm gonna pay I'm gonna make the payment I'm
just signing the request right now it's pinging those endpoints I don. I'm not logged in on this, so this is like a live
score. And there we go. I can see the receipt. I just made the payment on scale, 0.01 cent.
And I can see the transaction on the scale base Explorer so that is scale export to
payments working with cred to quantify reputation at scale amazing amazing no
I love it that's great and yeah anybody else this is live so anybody can go to the credit protocol slash try and, you know, make some requests, see your own credit score, see Vitalik's credit score, see his credit report.
I don't want to, I already don't like looking at my real credit score.
No, but Julian, thank you so much for hanging out.
Really, really appreciate it.
It looks like what you guys are building is really going to be able to help with the,
you know, the lending and borrowing, you know, aspect of the Web3 crypto ecosystem.
Before we let you go, anything further?
Yeah, I mean, well, just to plug, we're, you know, we're moving strongly into this
agentic reputation space.
So that's supporting ERC 8004.
So if there are other agent builders out there that are looking to understand the interactions they have with other agents,
we'd love to hear from them and build some great stuff together.
100%.
Awesome, Julian.
Thank you so much
and yeah excited to see where
cred goes keeps on going
thanks for having me guys
have a good one
cheers
well
Manel
episode 2
done and dusted
it's done
yeah some cool projects yeah um and yeah we're gonna excited
to continue learning about these you know incredible builders um by you man anything
else you want to say before we throw throw things away um yeah just say, like, just come visit SCALE and they can see some kids, check the Twitter
page.
Like you saw, we have a lot of different projects with amazing ideas.
And this is just the beginning, right? just um at the surface of x42 payments uh a genetic payments uh a 8004 um and i think in
three months four months they will have really really incredible projects big ones because yeah
for example this this guy cred um the guys from cred they can when they start making the integrations in the right places
i think they can be huge just like primer systems yeah 100 100 all righty well thank you all so
much for tuning in and joining us for another episode of the scale developers showcase we will
be back in two weeks for a new set of incredible partners
we hope to see you then bye bye
