Open Intents Framework: Solving Interop - Jim | LI.FI

Hello. Thanks for having me. Um, thank you for the ETH Belgrade organizers for having me on board and I was super excited to be here today to talk about something called the open intense framework. Um, you know, I know it's day three of a very long conference and you guys are probably mentally checked out. So I will try to keep this as high level and as entertaining as possible and I can guarantee you that this is probably going to be top five three five talks at the conference because this stuff is just genuinely groundbreaking like I cannot emphasize enough how interesting the open intense framework is not sarcasm so what is the open intense framework so the open intents framework is basically an open-source tooling kit built by a coalition of companies including Lei but also Ethereum Foundation, Open Zeppelin, Across, Hyperlane, and others uh to basically fix interoperability.

And so we're allowing for permissionless crosschain interoperability between any chains that exist now and are going to exist in the future as well. So basically trying to solve that cold start problem that we're seeing with a lot of new chains, a lot of roll-ups in our ecosystem. Uh but before I get started, uh let me tell you a little bit about myself and why I'm even qualified to talk about this stuff right now. Uh so my name is Jim. Uh I'm the head of product at Leifi.

Some of you guys may know me online as my handle is URX Jim, but please don't call me that out loud. It's I like to be called Jim. Um and previously I was a co-founder of a company called Catalyst where we built crosschain bridges. We actually got acquired by Leifi earlier this year. And so now I am at Lei uh still doing crosschain stuff because I love it.

Uh and then before that I was a product manager at a where I built a lot of crosschain stuff at a like portals and go and a lot of lens chain kind of p stuff as well and so definitely been doing a lot of crosschain things uh for the past few years. Uh and I've been in crypto for about seven years. is I started off working at Ripple as a product manager in 2017. Uh and now I'm here not working at Ripple, working in Ethereum or multi-chain. Uh and then lastly in the intros, uh what is Lei for some of you guys who don't know?

Uh Lei is one the company I work for, uh but two, it's a it's a bridge index aggregator. And so all these, you know, different tools that you can use to go to different chains and swap to different tokens, we basically aggregate and abstract all of that. And so if you have any token on one chain and you want to go to literally any other token on another chain, uh you can do that through Lei through our routing and through our execution engine. And uh we're pretty legit. You know, we've done over $30 billion volume to date.

So we know what we're talking about. And we're trusted by over 400 customers. Uh knowing uh and notable apps include things you guys probably already use like Phantom and MetaMask and Robin Hood and all that other good wallet stuff. But that's enough about intros. Let's talk about the good And so what is the problem that we have right now?

We have what I call the interrupt problem. And it's a problem that I think everyone in crypto is facing is that we have a lot of chains. Here are a subsection of them. It's not even all of them. I think there are over 350 chains on Coin Gecko and there are more chains launching each week.

We have things like rollup as a service infrastructures like Caldera and Conduit where you could literally spin up a new chain, new rollup just like a snap, right? Literally, I could do it right now if I had my computer and they allowed me a screen share. And so, if there are all these new chains launching each week and all these new app chains and app roll-ups and everything that everyone this conference has been talking about, really excited about keep on coming and there's more on the horizon. Uh, we got a problem on our hands, right guys? You know, we have these new chains.

It's super easy to make a new chain. What happens afterwards? What are you going to do? You just going to hang out on your own little chain? No.

Obviously, you want to have a connected economy, right? You want to be talking to all the other chains that you want to do business with. You want to go to base. You want to go to arbitum optimism. You even want to go beyond that, right?

You want to go to Salana. You want to go to the Cosmos ecosystem. You want to go move chains. I've made my points. But the issue is that there's no easy way to connect them together.

There's no easy way to do bridging because bridging is really, really hard. And not only is it really hard, but it's also a terrible user experience to have different wallets with different assets trying to move between different chains and it's just really painful and the user sees that and I think that's a huge blocker for mainstream adoption for crypto. And if we want to be anything to anyone as an industry and taken seriously, we need to solve this interop problem. And so the current solution right now, which is okay, is something called intents. Raise your hand if you know what intents are.

Okay, that's pretty good. That's that's more than I that's more than I thought outside of the little interop nerd bubble that I find myself in. Uh but as a primer for folks that don't know what intents are, intents are basically well there are many things but in the context of crosschain intents are basically a way for users to express what they want to do on different chains and basically offload the responsibility of that actual path finding and execution to a third party that we typically call a solver. And so here you can see that you know the origin chain is basically where the user finds themselves. They have some assets here.

They put it into escrow. So just a standard smart contract with maybe a time lock and some verification calls. And then you have this entity here called a solver that basically does all the magic, right? And so what a solver does very simplistically is that they have some assets on the chain that the user wants to go to and they basically just give it to the user. Obviously not for free.

They're going to be charging a little bit of money for that. But what's really cool about using intents is that now there's basically all these sophisticated actors like solvers who can take assets on one chain from a user and basically go to any destination chain they want, right? Uh not necessarily bounded to an n equals 1 context here or even n equals 2. Um, and so that's really helpful because now a user can have USDC on base, but they can perform actions or they can receive assets on all these other chains on Ethereum mainet, on arbitrum, on optimism, even on Salana and other ecosystems. And so that's really cool, right?

And that's what everyone's been really excited about and that's what everyone's been cheering about because they think quote interop is solved because we basically allow these solvers to just do the interopy thing for users right wrong and so there is a problem with intents unfortunately uh and I think the highle takeaway is that intents are really hard to build it's a huge barrier to entry to even have intent Hence exist on a certain chain. Obviously from the diagram before you could see that you need a solver, right? And a solvers are quite sophisticated and they hold inventory and they're basically market makers. And so it's not easy to be running a solver. And so you either have two choices in front of you if you're a chain that wants to have interop and have intents.

The first one is that you're relying on BD for integrations. And so basically, if you're a new chain, you have to go to these solvers and you have to beg them basically or or give them incentives, give them money uh to actually be on these chains, right? And actually hold inventory on these chains and solve routes for their users and everything. Uh and so that is obviously not permissionless. It's a huge barrier.

The second thing is that it's really complex to build and maintain your own solver, right? Let's say you don't want to sell your soul of the devil, you actually want to do it yourself. Um, it's really hard. You know, you have to do routing, you have to have inventory, you have to do rebalancing, all these other things. It's really hard.

And so what ends up happening is that each team that wants to build their own solver has to do it from scratch, right? There's no framework for how to build a solver. Everyone does it a little bit differently. Every chain is a little bit different. And so it ultimately becomes a really inefficient effort for the ecosystem as a whole.

And so we're gonna try to eliminate all that wasted work. And we're going to try to lower the barrier for intents to be able to be built and deployed on all these new chains. And we call that the open intents framework, the OAF. And again, this is a permissionless open-source framework for any chain. and any builder to deploy intents on their chain and have solvers move money and be aware of these intents all between between all these different chains.

And again, this is not something that we're building alone as Lei that, you know, I think defeats the purpose of building open source and building standards, but we're doing this as a coalition in tandem with the Ethereum Foundation and other builders in the space because we really want to fix this because we think it's obviously the biggest problem that we're facing in crypto right now. And so what is the open intense framework? So the open intense framework is basically a collection of modular modules of modules uh that allow builders to deploy different pieces of intents and to customize them based on all the little customizations and preferences that they have. Right? And the purpose of the open intents framework is that it's full stack.

It's end to end. And so there's no piece about intents that isn't covered in the OIF. And so any chain team that has no idea how to do crosschain at all can just use it off the shelf and be able to have intents from day one upon chain launch. And so these are the six components that we basically build for the OAF. And so at the top we have orderflow origination which is a a really fancy word for a UI or a website.

Uh but basically what we do is we allow an open- source website uh that anyone can fork and customize in order to you know issue intents and to receive intents. Uh the second thing is intent expression and escrow. I'll talk about that in a second. Uh the third one is an orderflow auction. And so after you know these intents are expressed uh basically it needs to go through a mechanism for these intents to be actually bid against uh by all these solvers.

U fourthly we have something called intent fulfillment. And so that is where solvers who've won the bid for an intent can actually execute it on chain. Fifthly we have something called crosschain validation. And so this is where after a solver has fulfilled an intent, uh, they're owed money, right? They need to get money somehow because they basically lent money to the user to perform this action.

And so there needs to be a way for them to uh show proof that they actually did that on the destination chain and then claim the assets on the origin chain. And then lastly, it's something called solver rebalancing. And so ultimately solvers hold money too and they need money in different places so that they can fulfill more intents. Um and so we build tooling that allows them to really easily do that. And so mind you these six pieces are what every solver does basically or every intent product does.

They have a website, they have intent expression and escrow. They have order flow auction. They have intent fulfillment blah blah blah blah blah. But they all build it themselves. And so there's no standards.

Everything is bespoke. Everything is built from scratch. And so obviously you can tell that's really inefficient. And so what we built with the OAF is that we've taken all these six pieces and we've basically done two things. One is that we've built an open-source version for all six of these things.

And the second thing that we've done is that we've allowed for the modularization of each piece such that there is customization on every level. So if you're a chain or if you're a builder team and you have all these little requirements that your chain has, it has super fast block speed or you want permission solving with certain market makers, you can do that, right? Like there's no stone that's left unturned. And I'm going to talk about this more in a second, but basically what we're trying to do with the OAF is that we're trying to build basically an open-source menu of every interop project. And so this is a research piece that uh I wrote with our head of research uh a couple months ago called the intent value chain.

Sorry, it's a it's a little small there, but basically we map out all the different pieces of how intents work. It's basically this thing right here. uh and we mapped out every project that's involved with it, right? So across and layer zero and stargates and one balance etc etc they all fit in here and so we're basically building a framework in which every project can contribute modules open source modules to customize anything right so you can build an open- source UI and you can fork it and you can customize how it looks u you can also build a chain abstracted wallet which I think many teams even at this conference were talking about and you plug that right into the open intents framework, right? And so that's that's pretty standard.

There are things that people are doing on the intent expression and escrow side. And so again, this is how a user expresses an intent and how they actually store money on chain such that they can even get authorized to issue an intent because then a solver knows that they have money and then they can actually fulfill an intent for them. Uh, and so things like ERC7683, raise your hand if you know what ERC7683 is. All right, no worries. It's basically a standard that across and unis swap wrote to standardize intent expression.

There's also things called the compact, which is basically another thing. I don't need to belver the point of all this nerdy stuff. Um, but that's something that we can customize, right? because when you're an interop project or you're you're a a chain, you actually have an a preference between these things once you get down to the nitty-gritty, although it may not seem like it right now, so I apologize. Uh, another thing uh that we customize and I kind of touched upon is orderflow auctions.

And so you can have a public or private auction. So that means any solver can participate in this order flow or it means only a permission set of solvers can participate and there's benefits to them. You can do things like a dodge auction. You can do things like a first come first serve and so basically like a like a gas war. Uh and there's you know minute benefits to each of these but they uh they accumulate over time and so these are preferences that people have.

Um intent fulfillment I think is a really big piece of the so like I said it's really hard to build a solver. Everyone builds it from scratch. And so we open sourced uh a solver implementation that can configure to all these pieces, right? And so um the really cool thing now is that anyone can use this off the shelf and just run their own solver and they don't have to build anything which is really helpful. Uh crosschain validation typically what happens right now with crosschain validation is that it's all bespoke.

they either use one oracle basically like chain link CCIP or maybe they use layer zero or wormhole etc etc. Um we have built a framework that allows for any crosschain validation to happen and there's actually been some work on the standard side. So there's something called um RIP7755. So rollup um improvement proposal 7755 you don't need to know the details of it. Uh there's another one called ERC7786.

It's basically uh allowing for the aggregation of all these different oracles. There's things called GMPs, general message protocols, uh like wormhole, like uh hyperlane, like layer zero that can arbitrarily pass any messages. Again, it doesn't really matter for the O of we can do anything, right? And people have an opinion on this because it's a security assumption they need to take on. And then lastly, we have solver rebalancing.

And so there this is really interesting because a really big barrier to running a solver besides the pure operations of it is actually the money operations of it right how do you actually move money to ensure that you have enough money to fulfill these intents on different chains and so this is where we're working with Everclear in order to do some really interesting netting and we're also working with another other liquidity layers um like nomial and something that we're building in house as well and oft layer Zero's omni chain funible token standard. Again, it's a lot of stuff, but basically the high level takeaway is that we can fight a lot of stuff in this baby. Okay, so very very excited that there is finally an open source framework that invites all these projects to basically become not competitors but cooperators in the goal of solving this huge interop problem. And so that's how it works in terms of all the nuts and bolts kind of. I don't have diagrams.

I could have added diagrams, but I can see that some eyes are glazing over, so I totally understand. Um, but if I'm a team or a chain developer, uh, how do I actually use the OIF, right? And so what you can do is you can go to the OAF repo and you can, you know, use our open source uh, deployer to deploy these OAF smart contracts uh, on this new chain that you want intense, right? And so if I build gym chain and it's a optimistic rollup using conduit, I can go to the OAF repo and I can basically deploy these open source contracts into my chain. Uh granted that's you know EVM equivalent but that's you know we have larger scope than EVM equivalency in the future.

Uh but for now EVM equivalency is is a requirement. Uh the second thing I do is I can customize any component to this right and so with the config I can update certain things. So I can use a 7683 for my intent expression. I can use a resource lock for escrow. I can have a uh first come first serve auction type, Oracle solver design, etc.

, etc. Like all the good stuff that I talked about that's obviously quite detailed. Um and then I fork the open source solver and I I run it myself, right? Uh using all the the straightforward documentation. Uh for folks that don't run want to run their own open source solver, we're also in the process of building a solver as a service.

And so basically an instance that you spin up and it's it's run by a third party that can actually run your solver for you. So even less of a barrier to be doing intents on this new chain. Uh and then I fork the open source UI. It's configured in a certain way such that it can detect new chains or sorry new intents on new chains. You just need to add the chain ID and the RPC for it.

Uh and what's the result? The result is that my new chain, my gym chain uh has intents now and I can connect it to base or optimism or ethereum mainet or anywhere and it just works and that's called solving interop ladies and gentlemen and so what does the give us? So the O of gives us permissionless chain deployment. Of course that is first and foremost the most important goal of the OAF. The second thing that it gives us is an offthe-shelf open-source modular infrastructure such that anyone can use it.

They can fork it. They can modify it and no one's ever building things from scratch. And so kind of just giving to the community and and allowing more people to run solvers, allowing more people to run intents with the hope that they would modify it as well and build new things that we've never seen before. And the last thing which I think is really beneficial for everyone in the audience is that we have one UI for all chains. This UI will have every chain will have every token and if there are new marginal chains being added, it will automatically add them as well.

And so you never need to figure out how you're gonna get different assets on different chains to another one because we have it soon. And so we're solving interop all these things that we had goals for that Vitalic has written down. We're going to solve them. We're going to solve day one chain support. We're going to do any chain any VM soon.

Again, not a hard requirement right now. anyone can run a solver. That's obviously a really important goal for the OAF. So, I highlighted it. But it's trust minimized.

It's open source. It's modular. It's chain abstracted. It's price competitive. It has all assets.

These are all things that we're solving with the OAF. And therefore, we're going to solve this interop problem. And the best part, and I mention again because it's really important, is that this is not a Lei effort. This is not something that Lei has concocted themselves and thought of. This is something that we're working with in tandem with other folks because if you're proposing a standard, but you're the only company using it, that's not a standard.

That's just something that you're using yourself. And so we're working with folks like the Ethereum Foundation, like Open Zeppelin across Hyperlane, Everclear, One Balance, you know, just all the logos here. And because we all realize that this is so painful, like genuinely like raise your hand if you think that the current bridging process sucks ass. Okay, like half the room, other half probably, you know, you guys are loving life uh with what you got. So, more power to you.

Uh but I'm one of those folks and I want to solve it and that's why I've dedicated a meaningful chunk of my adult life to trying to solve this problem. Jesus Christ. So, you can use the OAF today. It's a real thing. You can go to the OAF GitHub repo.

I mean, granted, you know, you have to be somewhat technical, uh but you can go to the OAF GitHub repo. You can deploy our contracts. Uh you can start issuing intents. Easy as that. I'm going to share the the repo link in a second.

Uh and to show you guys that it's real. I'm going to show you guys a demo right now. And I think we have a video that the lovely AV team will pull up and I can talk more about it. Cool. So, oh god, we're we're already starting.

So, basically, uh this is a this is our UI. It's called linttent.org. And what you could see here, obviously we're not, you know, UX UI designers here, but we can do things to issue an intent. And so here is where you manage your balance that we call the compact.

Here is where we're signing for an intent. And so here we want to sell three USDC and we want to buy three base USDC. That's an intent that was just signed earlier and was issued down here. And basically because there's no solvers right now, we're self-s solving it. So we filled it, executed a transaction, and now what we're doing here is that we are pasting the transaction hash to validate it.

Uh, and I think there should be something that pops up here. We'll see. Here we go. So basically, we're validating it. So it's already been filled and now it's getting validated uh that it's been filled using something called polymer which is one of the oracles that I mentioned earlier.

Uh and after it's been validated we can actually claim the funds that have been deposited on the 3 USDC on. So it's a little janky right now and I had to talk a little bit fast because I recorded a little fast. So sorry about that. Uh but you can go to it lintent.org.

You can follow these instructions here and you can do it themselves. You can see there's intents being issued and we're going to add more chains. Obviously, we're going to be on mainnet. Obviously, we're going to clean this stuff up, but for the sake of a demo, it's not bad, right? And I do apologize that it's a video that's going really fast.

I wasn't allowed to share my screen. Uh, not that I even would have wanted to because I probably would have messed it up anyway. Uh, but that is the video for it. So, again, lintent.org.

l.org uh to try it out. Cool. And I think we have a few more slides. Yes.

Thank you. So, I'll wrap up quickly. I know I'm at time. So, what's next for the OAF? And so, everything I talked about, we're finalizing all the smart contracts.

We're finalizing open source solver. We're making the UI a lot better and then eventually open source. But again, it's real and you can try it today. So, come get involved. is if this is something that's really cool for you, uh, scan the QR code.

It's going to go the the GitHub repo. You can read the contracts, you can read the solver, you can read the documentation, you can try the UI, it's all there for you. It's a real thing. I'm not just saying marketing stuff. Uh, and if this is something that interests you, if you want to get involved any way, again, it's an open source coalition.

Uh, DM me. I'm uh zeroxy on Twitter. I'm pretty responsive to DMs. Unless you're trying to scam me, then I will ignore you. Um, but yeah, that is all I got for you guys today.

And if you're interested in building the future of interoperability, um, scan this QR code or come ping me. Cool. Thank you. I think I have time for one question if the audience has a question. I know it was very dense, so I apologize.

Yeah. Hi. Hello, I have a question. What is your take on L2 and L3 solution right now? Do we really need them if the main liquidity is concentrated on top five blockchains?

Um, so your questions around um if we need something like because

no, if we need like L2L3 solutions that is launching every day. So all these networks if we really do need them.

Yeah. Yeah. So for for folks uh that that didn't hear. So the questions around you know there are five rollups that have you know line share of the majority of the liquidity and activity do we even need all these new marginal chains? Um so two answers to that.

So the first answer is that um even if we don't need them they're still coming right and so we have folks like that are still building new chains. We even have really interesting chains coming out like Mega ETH and Monad. Um but there's always the new marginal chain that's happening. There's a ton of new like stable coin rollups that are popping up due to all the stable coin craze right now and we need to connect them right the reality is that even if all the activity is happening on base and arbitrum etc we still need to connect all these otherwise it's a fragmented experience right it's like saying if there were new like countries in the world but no one ever flew to them because you know no one wants to go there basically so we need to at least provide infrastructure for it um and then secondly I I would say that um um we might need them in the future if we change to like an app rollup paradigm where every app or at least an ecosystem of apps are using their own block space. Uh again that's obviously been said for the past few years now.

We haven't really seen it in fruition outside of like Abstract and Sonium um and others. Uh but it might happen. But I think that's kind of my view on it. I think I'm at time. Thank you guys.