Rebuilding the web for agents — Liad Yosef, MCP Apps

Hi everyone. Hi everyone. Uh hello. Hello. So we're going to talk Uh hello. Hello. So we're going to talk Uh hello. Hello. So we're going to talk a little bit about the agentic web and a little bit about the agentic web and a little bit about the agentic web and more specifically what it means and how more specifically what it means and how more specifically what it means and how do we make the web ready for agents. Um do we make the web ready for agents. Um do we make the web ready for agents. Um one disclaimer um I built this talk one disclaimer um I built this talk one disclaimer um I built this talk yesterday so it might be outdated yesterday so it might be outdated yesterday so it might be outdated because uh things are moving really because uh things are moving really because uh things are moving really really fast in this uh in this space and really fast in this uh in this space and really fast in this uh in this space and um I think I need to introduce myself um I think I need to introduce myself um I think I need to introduce myself just to um to get some context. So I'm just to um to get some context. So I'm just to um to get some context. So I'm the co-creator and maintainer of a spec the co-creator and maintainer of a spec the co-creator and maintainer of a spec called MCP apps. MCP apps that's the called MCP apps. MCP apps that's the called MCP apps. MCP apps that's the underlying spec be behind uh CHP apps underlying spec be behind uh CHP apps underlying spec be behind uh CHP apps cloud apps copilot GitHub every cloud apps copilot GitHub every cloud apps copilot GitHub every chatbased app that you so that you see chatbased app that you so that you see chatbased app that you so that you see is based on the spec MCP apps in the MCP is based on the spec MCP apps in the MCP is based on the spec MCP apps in the MCP committee I'm also the co-founder of a committee I'm also the co-founder of a committee I'm also the co-founder of a company called Aura where we research company called Aura where we research company called Aura where we research the agentic human interactions and I the agentic human interactions and I the agentic human interactions and I built and led the agentic storefronts in built and led the agentic storefronts in built and led the agentic storefronts in Shopify so the topic of agentic web is Shopify so the topic of agentic web is Shopify so the topic of agentic web is really close to my heart um and a a really close to my heart um and a a really close to my heart um and a a little bit primer about MCP apps if little bit primer about MCP apps if little bit primer about MCP apps if you're not familiar. Um MCP apps are you're not familiar. Um MCP apps are you're not familiar. Um MCP apps are actually this is actually the spec that actually this is actually the spec that actually this is actually the spec that paved the way to Agentic Web. So MCP paved the way to Agentic Web. So MCP paved the way to Agentic Web. So MCP apps were released as a spec and as a apps were released as a spec and as a apps were released as a spec and as a standard few months ago um with the standard few months ago um with the standard few months ago um with the support of Claude first as the first support of Claude first as the first support of Claude first as the first client and then all clients followed.

client and then all clients followed. client and then all clients followed. And if you play with any chatbased app And if you play with any chatbased app And if you play with any chatbased app and you pulled any uh visualization and and you pulled any uh visualization and and you pulled any uh visualization and interaction UI layer from an MCP server, interaction UI layer from an MCP server, interaction UI layer from an MCP server, you probably used MCP apps. And the good you probably used MCP apps. And the good you probably used MCP apps. And the good thing about MCP apps is that everybody thing about MCP apps is that everybody thing about MCP apps is that everybody wins because if the servers or the wins because if the servers or the wins because if the servers or the providers can send UI chunks into the providers can send UI chunks into the providers can send UI chunks into the chats, then the apps gain their brand chats, then the apps gain their brand chats, then the apps gain their brand and identity. they get they get to keep and identity. they get they get to keep and identity. they get they get to keep their UI and not being reduced to a their UI and not being reduced to a their UI and not being reduced to a database or just textbased uh database or just textbased uh database or just textbased uh information. information. information. The users gain the trust and familiarity The users gain the trust and familiarity The users gain the trust and familiarity because if you uh ask JGPT, hey, book because if you uh ask JGPT, hey, book because if you uh ask JGPT, hey, book some hotels for me and you see some hotels for me and you see some hotels for me and you see Booking.com, then you know it's Booking.com, then you know it's Booking.com, then you know it's Booking.com, right? You know who you're Booking.com, right? You know who you're Booking.com, right? You know who you're interacting with and the host or the interacting with and the host or the interacting with and the host or the chats gain uh access to the world of chats gain uh access to the world of chats gain uh access to the world of capabilities. Um you know, a few months capabilities. Um you know, a few months capabilities. Um you know, a few months ago people ask, yeah, why won't OpenAI ago people ask, yeah, why won't OpenAI ago people ask, yeah, why won't OpenAI just build everything out of scratch? just build everything out of scratch? just build everything out of scratch? But open eye will not negotiate uh deals But open eye will not negotiate uh deals But open eye will not negotiate uh deals with hotels or support users that want with hotels or support users that want with hotels or support users that want to change seat in a venue. Um we need to change seat in a venue. Um we need to change seat in a venue. Um we need those services and MCP apps actually those services and MCP apps actually those services and MCP apps actually solve for this uh for this last mile solve for this uh for this last mile solve for this uh for this last mile interaction.

Yeah. So um uh this is the this is the Yeah. So um uh this is the this is the benefit that we get benefit that we get benefit that we get and MCP apps solved our last mile of and MCP apps solved our last mile of and MCP apps solved our last mile of interaction. Right. So what does it mean interaction. Right. So what does it mean interaction. Right. So what does it mean the last mile? Last mile is as agents the last mile? Last mile is as agents the last mile? Last mile is as agents become better, as models become better, become better, as models become better, become better, as models become better, they can do things more autonomously, they can do things more autonomously, they can do things more autonomously, but we are still in the end of the chain but we are still in the end of the chain but we are still in the end of the chain as humans. We still need to be able to as humans. We still need to be able to as humans. We still need to be able to choose the hotel or to view a 3D model choose the hotel or to view a 3D model choose the hotel or to view a 3D model or to choose a seat at the venue. all or to choose a seat at the venue. all or to choose a seat at the venue. all those last mile of interaction and that those last mile of interaction and that those last mile of interaction and that can change and um this is an example can change and um this is an example can change and um this is an example from uh um oh sorry yeah so this is from from uh um oh sorry yeah so this is from from uh um oh sorry yeah so this is from like cloud um APR on MCPF this is how it like cloud um APR on MCPF this is how it like cloud um APR on MCPF this is how it looks like right so when you um when looks like right so when you um when looks like right so when you um when you're embedding apps inside chats you you're embedding apps inside chats you you're embedding apps inside chats you get this unified experience of how UI get this unified experience of how UI get this unified experience of how UI feels inside the same context of a chat feels inside the same context of a chat feels inside the same context of a chat and as you can see in the same and as you can see in the same and as you can see in the same conversation you get data from booking conversation you get data from booking conversation you get data from booking you get data from all trails you can get you get data from all trails you can get you get data from all trails you can get data from everything you um uh you want data from everything you um uh you want data from everything you um uh you want to interact with. Right? So this is this to interact with. Right? So this is this to interact with. Right? So this is this is already happening. MCP apps is is already happening. MCP apps is is already happening. MCP apps is already widely supported across every uh already widely supported across every uh already widely supported across every uh chat agent except Gemini, but that's chat agent except Gemini, but that's chat agent except Gemini, but that's coming. Um and this is already coming. Um and this is already coming. Um and this is already happening. And the interesting part is happening. And the interesting part is happening. And the interesting part is that it brings us to what we call the that it brings us to what we call the that it brings us to what we call the agentic web. So agentic web. So agentic web. So what's what is the agentic web? The what's what is the agentic web? The what's what is the agentic web? The agentic web is not a web of agents. Uh agentic web is not a web of agents. Uh agentic web is not a web of agents. Uh it's not the web as we know it today.

it's not the web as we know it today. it's not the web as we know it today. It's a it's a shift. It's a change. It's It's a it's a shift. It's a change. It's It's a it's a shift. It's a change. It's a a change in in the way that we see a a change in in the way that we see a a change in in the way that we see websites and the way that we see websites and the way that we see websites and the way that we see browsers. Because up until now, we spent browsers. Because up until now, we spent browsers. Because up until now, we spent 20 years, two decades perfecting 20 years, two decades perfecting 20 years, two decades perfecting this experience. Meaning that if I want this experience. Meaning that if I want this experience. Meaning that if I want to somehow um do a project or fulfill a to somehow um do a project or fulfill a to somehow um do a project or fulfill a task or plan an anniversary, I have to task or plan an anniversary, I have to task or plan an anniversary, I have to open tabs in my browser and I have to go open tabs in my browser and I have to go open tabs in my browser and I have to go through these tabs and I have to convey through these tabs and I have to convey through these tabs and I have to convey my intent differently to each of those my intent differently to each of those my intent differently to each of those services, right? Meaning that if I just services, right? Meaning that if I just services, right? Meaning that if I just want to plan an anniversary, then I need want to plan an anniversary, then I need want to plan an anniversary, then I need to learn to learn to learn Google's UI and Amazon's UI and booking Google's UI and Amazon's UI and booking Google's UI and Amazon's UI and booking UI and the other booking UI and the UI and the other booking UI and the UI and the other booking UI and the other Amazon UI. All that just convey other Amazon UI. All that just convey other Amazon UI. All that just convey the same intent to those interfaces and the same intent to those interfaces and the same intent to those interfaces and every company perfected this user flow. every company perfected this user flow. every company perfected this user flow. But we don't need to do it anymore, But we don't need to do it anymore, But we don't need to do it anymore, right? We can just take these interfaces right? We can just take these interfaces right? We can just take these interfaces and break them into atoms because we and break them into atoms because we and break them into atoms because we don't need I don't need 99% of bookings don't need I don't need 99% of bookings don't need I don't need 99% of bookings dashboard. I don't need 99% of Airbnb's dashboard. I don't need 99% of Airbnb's dashboard. I don't need 99% of Airbnb's dashboard. I definitely don't need dashboard. I definitely don't need dashboard. I definitely don't need Jira's dashboard. But I do want to Jira's dashboard. But I do want to Jira's dashboard. But I do want to convey my intent to those services. So convey my intent to those services. So convey my intent to those services. So why can't I let my personal assistant why can't I let my personal assistant why can't I let my personal assistant use that? So if I have a personal use that? So if I have a personal use that? So if I have a personal assistant, I can just take these atoms assistant, I can just take these atoms assistant, I can just take these atoms and it composes these atoms and say, and it composes these atoms and say, and it composes these atoms and say, "Yeah, I see that you have an "Yeah, I see that you have an "Yeah, I see that you have an anniversary coming. This is this is the anniversary coming. This is this is the anniversary coming. This is this is the view from Google that says that you have view from Google that says that you have view from Google that says that you have an anniversary coming." So I can book I an anniversary coming." So I can book I an anniversary coming." So I can book I can I can buy things for you. I can book

can I can buy things for you. I can book can I can buy things for you. I can book your hotel. Now Claude knows me. So it your hotel. Now Claude knows me. So it your hotel. Now Claude knows me. So it it knows that I prefer to book hotels in it knows that I prefer to book hotels in it knows that I prefer to book hotels in uh in the nature, right? So it knows to uh in the nature, right? So it knows to uh in the nature, right? So it knows to pull the map from booking. I don't need pull the map from booking. I don't need pull the map from booking. I don't need to think about it. And what booking and to think about it. And what booking and to think about it. And what booking and Amazon and Google benefits from this is Amazon and Google benefits from this is Amazon and Google benefits from this is that they have this integration layer that they have this integration layer that they have this integration layer that they don't they don't need to that they don't they don't need to that they don't they don't need to develop because cloud has the context on develop because cloud has the context on develop because cloud has the context on for me, right? So booking doesn't need for me, right? So booking doesn't need for me, right? So booking doesn't need to develop integration with my calendar. to develop integration with my calendar. to develop integration with my calendar. Amazon doesn't need to develop Amazon doesn't need to develop Amazon doesn't need to develop integration with booking. So it's a integration with booking. So it's a integration with booking. So it's a win-win. This is the view that we're uh win-win. This is the view that we're uh win-win. This is the view that we're uh aiming for. And this is very different aiming for. And this is very different aiming for. And this is very different from what the things are going which is from what the things are going which is from what the things are going which is this which is basically browsing into this which is basically browsing into this which is basically browsing into different services just to stare at different services just to stare at different services just to stare at different text boxes that ask what I different text boxes that ask what I different text boxes that ask what I want to do there. But no one is going to want to do there. But no one is going to want to do there. But no one is going to do that in few in a year. No one is do that in few in a year. No one is do that in few in a year. No one is going to um browse to Amazon's agents going to um browse to Amazon's agents going to um browse to Amazon's agents and and Etsy's agents and Expedia's and and Etsy's agents and Expedia's and and Etsy's agents and Expedia's agent. I don't want to do that. I have agent. I don't want to do that. I have agent. I don't want to do that. I have my own personal assistant. I don't want my own personal assistant. I don't want my own personal assistant. I don't want I don't want to use your agent. I want I don't want to use your agent. I want I don't want to use your agent. I want to use my agent. Okay. Now, if we think to use my agent. Okay. Now, if we think to use my agent. Okay. Now, if we think about it, it means that websites as the about it, it means that websites as the about it, it means that websites as the source of truth are something that's source of truth are something that's source of truth are something that's going to go away because why would they going to go away because why would they going to go away because why would they need the website? Why would I need to need the website? Why would I need to need the website? Why would I need to open the tab? Why would they need the open the tab? Why would they need the open the tab? Why would they need the information that the website is working information that the website is working information that the website is working really hard to convey to me? And the really hard to convey to me? And the really hard to convey to me? And the immediate response to that is that yeah, immediate response to that is that yeah, immediate response to that is that yeah, we have browser agents. We have computer we have browser agents. We have computer we have browser agents. We have computer user agents. have very smart agents or user agents. have very smart agents or user agents. have very smart agents or assistants that can browse the web for assistants that can browse the web for assistants that can browse the web for us. But that doesn't make sense because

us. But that doesn't make sense because us. But that doesn't make sense because that that's that's a faster horses that that's that's a faster horses that that's that's a faster horses solution. Why would I let an agent that solution. Why would I let an agent that solution. Why would I let an agent that doesn't know anything about the human doesn't know anything about the human doesn't know anything about the human perception limitations work with uh perception limitations work with uh perception limitations work with uh filters and pagenations and sorting and filters and pagenations and sorting and filters and pagenations and sorting and all of those UIs that we spent decades all of those UIs that we spent decades all of those UIs that we spent decades perfecting for us um just for it to perfecting for us um just for it to perfecting for us um just for it to fulfill its task. So Google has a um fulfill its task. So Google has a um fulfill its task. So Google has a um Google is very bullish on the other end Google is very bullish on the other end Google is very bullish on the other end of the spectrum which is co- browsing of the spectrum which is co- browsing of the spectrum which is co- browsing and and websites and they have a and and websites and they have a and and websites and they have a protocol called or standard called protocol called or standard called protocol called or standard called WebMCP. WebMCP says okay instead of the WebMCP. WebMCP says okay instead of the WebMCP. WebMCP says okay instead of the agent taking screenshots of the websites agent taking screenshots of the websites agent taking screenshots of the websites and trying to figure out what's and trying to figure out what's and trying to figure out what's happening let's have the website expose happening let's have the website expose happening let's have the website expose some tools in the website JavaScript some tools in the website JavaScript some tools in the website JavaScript tools and the agent will walk with it. tools and the agent will walk with it. tools and the agent will walk with it. And you can see here an example of And you can see here an example of And you can see here an example of Gemini in Chrome that um buys stuff for Gemini in Chrome that um buys stuff for Gemini in Chrome that um buys stuff for me or uh finds uh deals for me as I me or uh finds uh deals for me as I me or uh finds uh deals for me as I browse the web, which is very nice, but browse the web, which is very nice, but browse the web, which is very nice, but it's very simple. What happens if we it's very simple. What happens if we it's very simple. What happens if we push Salesforce dashboard to Gemini on push Salesforce dashboard to Gemini on push Salesforce dashboard to Gemini on Chrome? Why would I want that? Why would Chrome? Why would I want that? Why would Chrome? Why would I want that? Why would I want to browse to Salesforce dashboard I want to browse to Salesforce dashboard I want to browse to Salesforce dashboard with Gemini to click buttons for me in a with Gemini to click buttons for me in a with Gemini to click buttons for me in a dashboard that I don't even want to use?

dashboard that I don't even want to use? dashboard that I don't even want to use? I mean, this is definitely not what we I mean, this is definitely not what we I mean, this is definitely not what we want. want. want. And the shift is that assistants are And the shift is that assistants are And the shift is that assistants are becoming our entry point to the web, becoming our entry point to the web, becoming our entry point to the web, right? We see it everywhere. Every major right? We see it everywhere. Every major right? We see it everywhere. Every major lab wants its app to be the everything lab wants its app to be the everything lab wants its app to be the everything app. And the shift is happening. And app. And the shift is happening. And app. And the shift is happening. And friends of mine that connect Jira's MCP friends of mine that connect Jira's MCP friends of mine that connect Jira's MCP to their IDs, they don't go to Jira's to their IDs, they don't go to Jira's to their IDs, they don't go to Jira's website anymore. Right? And my mother website anymore. Right? And my mother website anymore. Right? And my mother uses CHP for everything. If she could uses CHP for everything. If she could uses CHP for everything. If she could book a doctor's appointment using Chip, book a doctor's appointment using Chip, book a doctor's appointment using Chip, she won't go to that clinic's website she won't go to that clinic's website she won't go to that clinic's website anymore. and you have a long tail of anymore. and you have a long tail of anymore. and you have a long tail of websites that no one is going to see websites that no one is going to see websites that no one is going to see because it's much easier for me to because it's much easier for me to because it's much easier for me to connect to them using my personal connect to them using my personal connect to them using my personal assistant. So websites will become assistant. So websites will become assistant. So websites will become obsolete, browsers will become obsolete, obsolete, browsers will become obsolete, obsolete, browsers will become obsolete, and personal assistant will become our and personal assistant will become our and personal assistant will become our only gate to the web. And I know what it only gate to the web. And I know what it only gate to the web. And I know what it sounds like. It sounds like old men sounds like. It sounds like old men sounds like. It sounds like old men yelling at a lit literal cloud, right? I yelling at a lit literal cloud, right? I yelling at a lit literal cloud, right? I mean, um, yeah, they're not going to go mean, um, yeah, they're not going to go mean, um, yeah, they're not going to go completely obsolete, but we see it with completely obsolete, but we see it with completely obsolete, but we see it with the younger generation today. My the younger generation today. My the younger generation today. My 9-year-old 9-year-old 9-year-old goes to Chpt. It doesn't go to Google goes to Chpt. It doesn't go to Google goes to Chpt. It doesn't go to Google when it when it wants to search for when it when it wants to search for when it when it wants to search for stuff. I was uh I was traveling to stuff. I was uh I was traveling to stuff. I was uh I was traveling to Georgia, the country in Eastern Europe, Georgia, the country in Eastern Europe, Georgia, the country in Eastern Europe, and there was an older lady there that and there was an older lady there that and there was an older lady there that asked me to take a picture of her, like asked me to take a picture of her, like asked me to take a picture of her, like 70 year old, and on her phone, she had 70 year old, and on her phone, she had 70 year old, and on her phone, she had three apps, WhatsApp, camera, and Chip.

three apps, WhatsApp, camera, and Chip. three apps, WhatsApp, camera, and Chip. That's it. The shift is happening. Um, That's it. The shift is happening. Um, That's it. The shift is happening. Um, and very smart people like David Ker and very smart people like David Ker and very smart people like David Ker from Sentry said a year ago, I will take from Sentry said a year ago, I will take from Sentry said a year ago, I will take this bet against anyone who thinks this bet against anyone who thinks this bet against anyone who thinks websites are going to be obsolete in 25 websites are going to be obsolete in 25 websites are going to be obsolete in 25 years. That was one year ago. Few days years. That was one year ago. Few days years. That was one year ago. Few days ago, David Kramer published designing ago, David Kramer published designing ago, David Kramer published designing for agents which says we have to for agents which says we have to for agents which says we have to recognize that interaction with products recognize that interaction with products recognize that interaction with products at century will no longer be exclusively at century will no longer be exclusively at century will no longer be exclusively through our web application and we have through our web application and we have through our web application and we have to design for API first as a surface. So to design for API first as a surface. So to design for API first as a surface. So everything is going headless. Salesforce everything is going headless. Salesforce everything is going headless. Salesforce recently went headless which is very recently went headless which is very recently went headless which is very important because Salesforce main important because Salesforce main important because Salesforce main differentiator from its competitors is differentiator from its competitors is differentiator from its competitors is its UI. It's the way that it conveys its UI. It's the way that it conveys its UI. It's the way that it conveys everything to the user. But it went everything to the user. But it went everything to the user. But it went headless. Cloudflow went headless. headless. Cloudflow went headless. headless. Cloudflow went headless. Sentry went headless. Um this is a a Sentry went headless. Um this is a a Sentry went headless. Um this is a a tweet by Cloudfl CEO saying that agents tweet by Cloudfl CEO saying that agents tweet by Cloudfl CEO saying that agents traffic surpassed uh human traffic in traffic surpassed uh human traffic in traffic surpassed uh human traffic in the web. And we have this spectrum of the web. And we have this spectrum of the web. And we have this spectrum of interaction because if I if my agent is interaction because if I if my agent is interaction because if I if my agent is autonomous, I don't need this last mile autonomous, I don't need this last mile autonomous, I don't need this last mile of UI, right? If I have an open claw, I of UI, right? If I have an open claw, I of UI, right? If I have an open claw, I just send and I just say yeah organize just send and I just say yeah organize just send and I just say yeah organize my emails. It comes back said good. But my emails. It comes back said good. But my emails. It comes back said good. But if I want to book a hotel for my if I want to book a hotel for my if I want to book a hotel for my honeymoon, I probably need this less honeymoon, I probably need this less honeymoon, I probably need this less mile of interaction. So we call it the mile of interaction. So we call it the mile of interaction. So we call it the nearly headless web, right? It's not nearly headless web, right? It's not nearly headless web, right? It's not completely headless. Um the agents will completely headless. Um the agents will completely headless. Um the agents will interact with your web head with your um interact with your web head with your um interact with your web head with your um website headlessly, but then it will website headlessly, but then it will website headlessly, but then it will bring back these UI resources for you.

bring back these UI resources for you. bring back these UI resources for you. So user experience as we know it is So user experience as we know it is So user experience as we know it is being split. Now it's the agent being split. Now it's the agent being split. Now it's the agent experience of your website plus how a experience of your website plus how a experience of your website plus how a user experiences the agent that user experiences the agent that user experiences the agent that experiences your website. But 90% is experiences your website. But 90% is experiences your website. But 90% is going to rely on agent experiences which going to rely on agent experiences which going to rely on agent experiences which means your website needs to be agent means your website needs to be agent means your website needs to be agent ready for everything. It needs to be ready for everything. It needs to be ready for everything. It needs to be agent ready for your agent. If I'm agent ready for your agent. If I'm agent ready for your agent. If I'm booking I need to make bookings data booking I need to make bookings data booking I need to make bookings data ready for bookings agent. I need to make ready for bookings agent. I need to make ready for bookings agent. I need to make it ready for my customers chpt or my it ready for my customers chpt or my it ready for my customers chpt or my customers open claw and I need I need to customers open claw and I need I need to customers open claw and I need I need to be ready for my human customer who be ready for my human customer who be ready for my human customer who doesn't have an agent but still wants to doesn't have an agent but still wants to doesn't have an agent but still wants to browse to booking.com right so I need to browse to booking.com right so I need to browse to booking.com right so I need to to be agent ready and what we find found to be agent ready and what we find found to be agent ready and what we find found out is that agent ready means a lot of out is that agent ready means a lot of out is that agent ready means a lot of things so we heard a lot about AEO SEO things so we heard a lot about AEO SEO things so we heard a lot about AEO SEO geo how to get discovered but discover geo how to get discovered but discover geo how to get discovered but discover discovery is only the first the first discovery is only the first the first discovery is only the first the first step because once an agent knows about step because once an agent knows about step because once an agent knows about you it still needs to know what you are you it still needs to know what you are you it still needs to know what you are and how to interact with you and how and how to interact with you and how and how to interact with you and how authenticate to you and how to authenticate to you and how to authenticate to you and how to headlessly pay you, right? And actually headlessly pay you, right? And actually headlessly pay you, right? And actually when when we built uh it's an anecdote, when when we built uh it's an anecdote, when when we built uh it's an anecdote, but when we built analytics for our but when we built analytics for our but when we built analytics for our product, we asked Cloud Code, what's the product, we asked Cloud Code, what's the product, we asked Cloud Code, what's the best analytic service? And it best analytic service? And it best analytic service? And it recommended Postto, which is an analytic recommended Postto, which is an analytic recommended Postto, which is an analytic service. And we said no, we prefer service. And we said no, we prefer service. And we said no, we prefer Mixpanel because we know Mixpanel. We Mixpanel because we know Mixpanel. We Mixpanel because we know Mixpanel. We worked with Mixpanel for a decade, we worked with Mixpanel for a decade, we worked with Mixpanel for a decade, we know how to work with it. And Cloud Code know how to work with it. And Cloud Code know how to work with it. And Cloud Code insisted on Postto because it said insisted on Postto because it said insisted on Postto because it said Postto has better MCP and API and I can Postto has better MCP and API and I can Postto has better MCP and API and I can integrate to it better. So we don't have integrate to it better. So we don't have integrate to it better. So we don't have brand loyalty, right? We went with brand loyalty, right? We went with brand loyalty, right? We went with postto hog if that's what cloud code postto hog if that's what cloud code postto hog if that's what cloud code recommend but it made us think about recommend but it made us think about recommend but it made us think about mixpan mixpan spent a decade perfecting mixpan mixpan spent a decade perfecting mixpan mixpan spent a decade perfecting their UIX and developer experience we

their UIX and developer experience we their UIX and developer experience we just left it just because cloud code just left it just because cloud code just left it just because cloud code prefers posto and it will always happen prefers posto and it will always happen prefers posto and it will always happen her for example if it uses your product her for example if it uses your product her for example if it uses your product and it needs to spin up a browser to use and it needs to spin up a browser to use and it needs to spin up a browser to use your product it will remember that and your product it will remember that and your product it will remember that and next time it won't go to your product next time it won't go to your product next time it won't go to your product anymore right so at aura which is an anymore right so at aura which is an anymore right so at aura which is an agentic web research lab we started agentic web research lab we started agentic web research lab we started researching uh we raised some raised researching uh we raised some raised researching uh we raised some raised some capital and we started researching some capital and we started researching some capital and we started researching what does it mean for the web to be what does it mean for the web to be what does it mean for the web to be ready for agents and we have these um ready for agents and we have these um ready for agents and we have these um readiness benchmarks which is readiness benchmarks which is readiness benchmarks which is interesting so you can run it's free you interesting so you can run it's free you interesting so you can run it's free you can go to aura.ai AI you can run any can go to aura.ai AI you can run any can go to aura.ai AI you can run any website you get this score and website you get this score and website you get this score and benchmarks according to a lot of benchmarks according to a lot of benchmarks according to a lot of protocols and best practices and you protocols and best practices and you protocols and best practices and you have agentic feedback. So the agent have agentic feedback. So the agent have agentic feedback. So the agent actually returns the feedback about your actually returns the feedback about your actually returns the feedback about your uh your website uh your website uh your website and we have this leaderboard of how and we have this leaderboard of how and we have this leaderboard of how companies and products rank according to companies and products rank according to companies and products rank according to this benchmark and we started mapping this benchmark and we started mapping this benchmark and we started mapping the web and everything was nice. the web and everything was nice. the web and everything was nice. But But But we hit one insight or one uh unexpected we hit one insight or one uh unexpected we hit one insight or one uh unexpected result. We found out result. We found out result. We found out anyone here heard about lm.txt txt anyone here heard about lm.txt txt anyone here heard about lm.txt txt lm.txt that's like the de facto standard lm.txt that's like the de facto standard lm.txt that's like the de facto standard to be agent ready. You say yeah if you to be agent ready. You say yeah if you to be agent ready. You say yeah if you if your website publish an lm.txt agent if your website publish an lm.txt agent if your website publish an lm.txt agent know how to interact with you um and you know how to interact with you um and you know how to interact with you um and you have o.md and pricing.mmd and x42 and a have o.md and pricing.mmd and x42 and a have o.md and pricing.mmd and x42 and a lot of standards and we found out that lot of standards and we found out that lot of standards and we found out that almost 50% of the website that we tested almost 50% of the website that we tested almost 50% of the website that we tested that we ran published lms.txt

that we ran published lms.txt that we ran published lms.txt but none of the agents that we ran on but none of the agents that we ran on but none of the agents that we ran on this website actually used llm.txt txt. this website actually used llm.txt txt. this website actually used llm.txt txt. Actually, almost all the agents went Actually, almost all the agents went Actually, almost all the agents went straight to the docs page and then the straight to the docs page and then the straight to the docs page and then the homepage. And the 40% that did use homepage. And the 40% that did use homepage. And the 40% that did use lm.txt use it only because the docs lm.txt use it only because the docs lm.txt use it only because the docs pointed out that there's a file called pointed out that there's a file called pointed out that there's a file called nmst txt that they need to to use. So nmst txt that they need to to use. So nmst txt that they need to to use. So then it hit us. We said it doesn't make then it hit us. We said it doesn't make then it hit us. We said it doesn't make a lot of sense for us as humans to a lot of sense for us as humans to a lot of sense for us as humans to define to agents what they need. No one define to agents what they need. No one define to agents what they need. No one is doing it anymore. Even OpenAI, they is doing it anymore. Even OpenAI, they is doing it anymore. Even OpenAI, they don't publish best practices for tools don't publish best practices for tools don't publish best practices for tools anymore because they say that every time anymore because they say that every time anymore because they say that every time they publish best practices, the models they publish best practices, the models they publish best practices, the models become better and these practices become become better and these practices become become better and these practices become obsolete. Six months ago, the best obsolete. Six months ago, the best obsolete. Six months ago, the best practice for an MCP server was yeah uh practice for an MCP server was yeah uh practice for an MCP server was yeah uh have three paragraph descriptions so have three paragraph descriptions so have three paragraph descriptions so agents will know how to interact with agents will know how to interact with agents will know how to interact with you and now three lines are enough. So you and now three lines are enough. So you and now three lines are enough. So best practices become we need the agents best practices become we need the agents best practices become we need the agents to define what the agent need, right? We to define what the agent need, right? We to define what the agent need, right? We need agents feedback on this website. So need agents feedback on this website. So need agents feedback on this website. So we built Aura's journey and Aura's we built Aura's journey and Aura's we built Aura's journey and Aura's journey that's a really cool that's also journey that's a really cool that's also journey that's a really cool that's also free. You can go to journey.ora.ai and free. You can go to journey.ora.ai and free. You can go to journey.ora.ai and you can run on any website any intent you can run on any website any intent you can run on any website any intent with any agent and see the path of the with any agent and see the path of the with any agent and see the path of the agent as it tries to interact with the agent as it tries to interact with the agent as it tries to interact with the website. So for example atio.com website. So for example atio.com website. So for example atio.com choose cl codec code and we run it on choose cl codec code and we run it on choose cl codec code and we run it on the website and you can see in real time the website and you can see in real time the website and you can see in real time how the agent goes and what it tries to how the agent goes and what it tries to how the agent goes and what it tries to look for in the website and we we did it look for in the website and we we did it look for in the website and we we did it tens of thousands of times just to tens of thousands of times just to tens of thousands of times just to understand what agents really look for

understand what agents really look for understand what agents really look for when they look when when they try to when they look when when they try to when they look when when they try to interact with with websites. interact with with websites. interact with with websites. And the cool thing is not running just And the cool thing is not running just And the cool thing is not running just one harness one harness one harness is running multiple harnesses. You see is running multiple harnesses. You see is running multiple harnesses. You see here cloud code and Eve which is here cloud code and Eve which is here cloud code and Eve which is Versel's harness um and CH GPT on the Versel's harness um and CH GPT on the Versel's harness um and CH GPT on the same website with the same intent. Okay. same website with the same intent. Okay. same website with the same intent. Okay. So this is cloud code. This is haiku. So this is cloud code. This is haiku. So this is cloud code. This is haiku. This is Eve. See how different the agent This is Eve. See how different the agent This is Eve. See how different the agent journey looks like. And this is CHP. So journey looks like. And this is CHP. So journey looks like. And this is CHP. So CHP could find results better. See you CHP could find results better. See you CHP could find results better. See you can just see the the the journey across can just see the the the journey across can just see the the the journey across across the website. And we need to across the website. And we need to across the website. And we need to understand why. We need to understand understand why. We need to understand understand why. We need to understand why does it happen? Why websites publish why does it happen? Why websites publish why does it happen? Why websites publish o.md files but agent don't look for o.md o.md files but agent don't look for o.md o.md files but agent don't look for o.md files and what do they look for? So in files and what do they look for? So in files and what do they look for? So in aura um you can actually uh go to any aura um you can actually uh go to any aura um you can actually uh go to any business business business question. So business business business question. So business business business question. So for any domain you have like business for any domain you have like business for any domain you have like business goals or or questions and you can see goals or or questions and you can see goals or or questions and you can see the paths that the agents are taking. the paths that the agents are taking. the paths that the agents are taking. And the uh interesting part of it is And the uh interesting part of it is And the uh interesting part of it is that okay now we know how agents that okay now we know how agents that okay now we know how agents interact with website. What's the next interact with website. What's the next interact with website. What's the next big big milestone for the agentic web big big milestone for the agentic web big big milestone for the agentic web and for those of you who heard the the and for those of you who heard the the and for those of you who heard the the previous talk? Um it's discovery, right?

previous talk? Um it's discovery, right? previous talk? Um it's discovery, right? But it's not the discovery that we think But it's not the discovery that we think But it's not the discovery that we think of. It's not SEO, it's not GEO, it's not of. It's not SEO, it's not GEO, it's not of. It's not SEO, it's not GEO, it's not AEO because we have to remember that AEO because we have to remember that AEO because we have to remember that every revolution came with its discovery every revolution came with its discovery every revolution came with its discovery layer, right? The web revolution came layer, right? The web revolution came layer, right? The web revolution came with search, um mobile came with app with search, um mobile came with app with search, um mobile came with app stores, social came with feeds. stores, social came with feeds. stores, social came with feeds. What is the discovery layer for aic What is the discovery layer for aic What is the discovery layer for aic resources? What is the discovery layer resources? What is the discovery layer resources? What is the discovery layer for MCPS? What is the discovery layer for MCPS? What is the discovery layer for MCPS? What is the discovery layer for open API.json? What do we even look for open API.json? What do we even look for open API.json? What do we even look for? I mean, maybe Airbnb has a better for? I mean, maybe Airbnb has a better for? I mean, maybe Airbnb has a better MCP than than Booking.com, but MCP than than Booking.com, but MCP than than Booking.com, but Booking.com has better API than Airbnb. Booking.com has better API than Airbnb. Booking.com has better API than Airbnb. So, how do we do it? Um, web search, So, how do we do it? Um, web search, So, how do we do it? Um, web search, classic web search, it's not enough classic web search, it's not enough classic web search, it's not enough because um it it's based on human SEO because um it it's based on human SEO because um it it's based on human SEO and human uh page rank and it's not and human uh page rank and it's not and human uh page rank and it's not flexible enough. custom registries like flexible enough. custom registries like flexible enough. custom registries like per agents or per chats, they're not per agents or per chats, they're not per agents or per chats, they're not enough because they're closed and enough because they're closed and enough because they're closed and they're uh um and and they require every they're uh um and and they require every they're uh um and and they require every app to to submit itself to these reg app to to submit itself to these reg app to to submit itself to these reg registries and a central registry like registries and a central registry like registries and a central registry like an MCP registry. That's not enough an MCP registry. That's not enough an MCP registry. That's not enough because who will do the curation? What's because who will do the curation? What's because who will do the curation? What's the governance model? How do we decide the governance model? How do we decide the governance model? How do we decide which resource comes to that registry?

which resource comes to that registry? which resource comes to that registry? There are emerging standards around it. There are emerging standards around it. There are emerging standards around it. There's AI catalog.json JSON which is a There's AI catalog.json JSON which is a There's AI catalog.json JSON which is a standard by entropic openai, Google, MCP standard by entropic openai, Google, MCP standard by entropic openai, Google, MCP and A2A which standardizes how a website and A2A which standardizes how a website and A2A which standardizes how a website exposes itself to agents and there's the exposes itself to agents and there's the exposes itself to agents and there's the agentic resource discovery standard agentic resource discovery standard agentic resource discovery standard which is by all these companies and more which is by all these companies and more which is by all these companies and more which is basically standardized how uh a which is basically standardized how uh a which is basically standardized how uh a discovery layer or or a directory uh discovery layer or or a directory uh discovery layer or or a directory uh expose itself to agents. So we built or expose itself to agents. So we built or expose itself to agents. So we built or directory right because we're a research directory right because we're a research directory right because we're a research lab for the agentic web. We built this lab for the agentic web. We built this lab for the agentic web. We built this and in or directory we take all the and in or directory we take all the and in or directory we take all the domains that we scanned or scanned the domains that we scanned or scanned the domains that we scanned or scanned the results and we put it in a directory results and we put it in a directory results and we put it in a directory that agents can actually query right um that agents can actually query right um that agents can actually query right um we expose the AI catalog.json JSON we expose the AI catalog.json JSON we expose the AI catalog.json JSON files. So you can see here for example files. So you can see here for example files. So you can see here for example for monday.com for monday.com for monday.com you can see that we generate this JSON you can see that we generate this JSON you can see that we generate this JSON file that basically tells agents yeah file that basically tells agents yeah file that basically tells agents yeah this is the MCP server for Monday this this is the MCP server for Monday this this is the MCP server for Monday this is the API server from monday.com and is the API server from monday.com and is the API server from monday.com and you can just the agent can just query you can just the agent can just query you can just the agent can just query that right so for every domain it can that right so for every domain it can that right so for every domain it can just goes to go to aic catalog.json JSON just goes to go to aic catalog.json JSON just goes to go to aic catalog.json JSON and we expose the registry the directory and we expose the registry the directory and we expose the registry the directory itself. So for every entry on the itself. So for every entry on the itself. So for every entry on the directory directory directory we can uh we can tell the agent yeah for we can uh we can tell the agent yeah for we can uh we can tell the agent yeah for example versel these are the agentic example versel these are the agentic example versel these are the agentic resources um and this is and this is how resources um and this is and this is how resources um and this is and this is how you access them and obviously this you access them and obviously this you access them and obviously this directory is fully ard or aentic directory is fully ard or aentic directory is fully ard or aentic resource discovery compliant. So any resource discovery compliant. So any resource discovery compliant. So any agent can just query that directory with agent can just query that directory with agent can just query that directory with any with any query. Right?

any with any query. Right? any with any query. Right? So this is uh this is aura. directory uh So this is uh this is aura. directory uh So this is uh this is aura. directory uh if you want to take a look all of it is if you want to take a look all of it is if you want to take a look all of it is part of the aura multiverse journey part of the aura multiverse journey part of the aura multiverse journey directory and the ranker. And we found directory and the ranker. And we found directory and the ranker. And we found out some very other interesting out some very other interesting out some very other interesting insights. For example being agent ready insights. For example being agent ready insights. For example being agent ready and being human accessible is very and being human accessible is very and being human accessible is very similar because LLMs when they come to similar because LLMs when they come to similar because LLMs when they come to your website they're like um users or your website they're like um users or your website they're like um users or people with vision disabilities. they people with vision disabilities. they people with vision disabilities. they don't see your website. They need other don't see your website. They need other don't see your website. They need other signals to understand how to walk with signals to understand how to walk with signals to understand how to walk with your website. And making your website your website. And making your website your website. And making your website human accessible helps agent human accessible helps agent human accessible helps agent accessibility and vice vice versa. accessibility and vice vice versa. accessibility and vice vice versa. So to wrap up, these are MCP apps. This So to wrap up, these are MCP apps. This So to wrap up, these are MCP apps. This is what I've been working on in the past is what I've been working on in the past is what I've been working on in the past few months. And this was the last piece few months. And this was the last piece few months. And this was the last piece for the agentic web. This actually for the agentic web. This actually for the agentic web. This actually brought the agentic web. And now agents brought the agentic web. And now agents brought the agentic web. And now agents start to roam the web. And agents need start to roam the web. And agents need start to roam the web. And agents need different things. They need different different things. They need different different things. They need different roads. They need different infras. But roads. They need different infras. But roads. They need different infras. But we don't want to rebuild the web for we don't want to rebuild the web for we don't want to rebuild the web for agents. We we want to make the web agent agents. We we want to make the web agent agents. We we want to make the web agent accessible. So let's just make sure the accessible. So let's just make sure the accessible. So let's just make sure the web is prepared for when agents are web is prepared for when agents are web is prepared for when agents are coming.

Thank you very much. Thank you very much. [applause]