WEBVTT

NOTE Sentence-level transcript of https://www.youtube.com/watch?v=waI44NP1abk

NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/waI44NP1abk.html. A cue ends where the next begins, or 2 s after its last word.

s1
00:00:01.309 --> 00:00:03.309
[music]

s2
00:00:12.480 --> 00:00:14.240
Hi everyone.

s3
00:00:14.240 --> 00:00:15.120
Uh hello.

s4
00:00:15.120 --> 00:00:16.160
Hello.

s5
00:00:16.160 --> 00:00:23.680
So we're going to talk a little bit about the agentic web and more specifically what it means and how do we make the web ready for agents.

s6
00:00:23.680 --> 00:00:36.960
Um one disclaimer um I built this talk yesterday so it might be outdated because uh things are moving really really fast in this uh in this space and um I think I need to introduce myself just to um to get some context.

s7
00:00:36.960 --> 00:00:40.800
So I'm the co-creator and maintainer of a spec called MCP apps.

s8
00:00:40.800 --> 00:00:59.840
MCP apps that's the underlying spec be behind uh CHP apps cloud apps copilot GitHub every chatbased app that you so that you see is based on the spec MCP apps in the MCP committee I'm also the co-founder of a company called Aura where we research the agentic

s9
00:00:59.840 --> 00:01:13.119
human interactions and I built and led the agentic storefronts in Shopify so the topic of agentic web is really close to my heart um and a a little bit primer about MCP apps if you're not familiar.

s10
00:01:13.119 --> 00:01:18.159
Um MCP apps are actually this is actually the spec that paved the way to Agentic Web.

s11
00:01:18.159 --> 00:01:28.479
So MCP apps were released as a spec and as a standard few months ago um with the support of Claude first as the first client and then all clients followed.

s12
00:01:28.479 --> 00:01:40.400
And if you play with any chatbased app and you pulled any uh visualization and interaction UI layer from an MCP server, you probably used MCP apps.

s13
00:01:40.400 --> 00:01:55.520
And the good thing about MCP apps is that everybody wins because if the servers or the providers can send UI chunks into the chats, then the apps gain their brand and identity.

s14
00:01:55.520 --> 00:02:03.119
they get they get to keep their UI and not being reduced to a database or just textbased uh information.

s15
00:02:03.119 --> 00:02:11.280
The users gain the trust and familiarity because if you uh ask JGPT, hey, book some hotels for me and you see Booking.com, then you know it's Booking.com, right?

s16
00:02:11.280 --> 00:02:19.360
You know who you're interacting with and the host or the chats gain uh access to the world of capabilities.

s17
00:02:19.360 --> 00:02:24.959
Um you know, a few months ago people ask, yeah, why won't OpenAI just build everything out of scratch?

s18
00:02:24.959 --> 00:02:32.319
But open eye will not negotiate uh deals with hotels or support users that want to change seat in a venue.

s19
00:02:32.319 --> 00:02:39.280
Um we need those services and MCP apps actually solve for this uh for this last mile interaction.

s20
00:02:41.680 --> 00:02:42.640
Yeah.

s21
00:02:42.640 --> 00:02:51.120
So um uh this is the this is the benefit that we get and MCP apps solved our last mile of interaction.

s22
00:02:51.120 --> 00:02:51.519
Right.

s23
00:02:51.519 --> 00:02:53.360
So what does it mean the last mile?

s24
00:02:53.360 --> 00:03:03.280
Last mile is as agents become better, as models become better, they can do things more autonomously, but we are still in the end of the chain as humans.

s25
00:03:03.280 --> 00:03:11.200
We still need to be able to choose the hotel or to view a 3D model or to choose a seat at the venue.

s26
00:03:11.200 --> 00:03:29.120
all those last mile of interaction and that can change and um this is an example from uh um oh sorry yeah so this is from like cloud um APR on MCPF this is how it looks like right so when you um when you're embedding apps inside chats

s27
00:03:29.120 --> 00:03:43.120
you get this unified experience of how UI feels inside the same context of a chat and as you can see in the same conversation you get data from booking you get data from all trails you can get data from everything you um

s28
00:03:43.120 --> 00:03:44.879
uh you want to interact with.

s29
00:03:44.879 --> 00:03:45.360
Right?

s30
00:03:45.360 --> 00:03:47.360
So this is this is already happening.

s31
00:03:47.360 --> 00:03:53.920
MCP apps is already widely supported across every uh chat agent except Gemini, but that's coming.

s32
00:03:53.920 --> 00:03:56.480
Um and this is already happening.

s33
00:03:56.480 --> 00:04:02.480
And the interesting part is that it brings us to what we call the agentic web.

s34
00:04:02.480 --> 00:04:05.920
So what's what is the agentic web?

s35
00:04:05.920 --> 00:04:07.760
The agentic web is not a web of agents.

s36
00:04:07.760 --> 00:04:09.920
Uh it's not the web as we know it today.

s37
00:04:09.920 --> 00:04:11.680
It's a it's a shift.

s38
00:04:11.680 --> 00:04:12.560
It's a change.

s39
00:04:12.560 --> 00:04:17.759
It's a a change in in the way that we see websites and the way that we see browsers.

s40
00:04:17.759 --> 00:04:26.320
Because up until now, we spent 20 years, two decades perfecting this experience.

s41
00:04:26.320 --> 00:04:39.120
Meaning that if I want to somehow um do a project or fulfill a task or plan an anniversary, I have to open tabs in my browser and I have to go through these tabs and I have to convey my intent

s42
00:04:39.120 --> 00:04:41.919
differently to each of those services, right?

s43
00:04:41.919 --> 00:04:53.840
Meaning that if I just want to plan an anniversary, then I need to learn Google's UI and Amazon's UI and booking UI and the other booking UI and the other Amazon UI.

s44
00:04:53.840 --> 00:05:00.960
All that just convey the same intent to those interfaces and every company perfected this user flow.

s45
00:05:00.960 --> 00:05:02.639
But we don't need to do it anymore, right?

s46
00:05:02.639 --> 00:05:11.759
We can just take these interfaces and break them into atoms because we don't need I don't need 99% of bookings dashboard.

s47
00:05:11.759 --> 00:05:14.479
I don't need 99% of Airbnb's dashboard.

s48
00:05:14.479 --> 00:05:17.199
I definitely don't need Jira's dashboard.

s49
00:05:17.199 --> 00:05:20.720
But I do want to convey my intent to those services.

s50
00:05:20.720 --> 00:05:25.600
So why can't I let my personal assistant use that?

s51
00:05:25.600 --> 00:05:33.039
So if I have a personal assistant, I can just take these atoms and it composes these atoms and say, "Yeah, I see that you have an anniversary coming.

s52
00:05:33.039 --> 00:05:36.800
This is this is the view from Google that says that you have an anniversary coming."

s53
00:05:36.800 --> 00:05:39.440
So I can book I can I can buy things for you.

s54
00:05:39.440 --> 00:05:40.560
I can book your hotel.

s55
00:05:40.560 --> 00:05:42.560
Now Claude knows me.

s56
00:05:42.560 --> 00:05:47.120
So it it knows that I prefer to book hotels in uh in the nature, right?

s57
00:05:47.120 --> 00:05:49.199
So it knows to pull the map from booking.

s58
00:05:49.199 --> 00:05:50.880
I don't need to think about it.

s59
00:05:50.880 --> 00:06:01.280
And what booking and Amazon and Google benefits from this is that they have this integration layer that they don't they don't need to develop because cloud has the context on for me, right?

s60
00:06:01.280 --> 00:06:03.600
So booking doesn't need to develop integration with my calendar.

s61
00:06:03.600 --> 00:06:06.319
Amazon doesn't need to develop integration with booking.

s62
00:06:06.319 --> 00:06:07.680
So it's a win-win.

s63
00:06:07.680 --> 00:06:10.880
This is the view that we're uh aiming for.

s64
00:06:10.880 --> 00:06:25.600
And this is very different from what the things are going which is this which is basically browsing into different services just to stare at different text boxes that ask what I want to do there.

s65
00:06:25.600 --> 00:06:28.560
But no one is going to do that in few in a year.

s66
00:06:28.560 --> 00:06:34.000
No one is going to um browse to Amazon's agents and and Etsy's agents and Expedia's agent.

s67
00:06:34.000 --> 00:06:34.639
I don't want to do that.

s68
00:06:34.639 --> 00:06:36.160
I have my own personal assistant.

s69
00:06:36.160 --> 00:06:39.120
I don't want I don't want to use your agent.

s70
00:06:39.120 --> 00:06:41.360
I want to use my agent.

s71
00:06:41.360 --> 00:06:42.400
Okay.

s72
00:06:42.400 --> 00:06:53.199
Now, if we think about it, it means that websites as the source of truth are something that's going to go away because why would they need the website?

s73
00:06:53.199 --> 00:06:55.120
Why would I need to open the tab?

s74
00:06:55.120 --> 00:07:00.160
Why would they need the information that the website is working really hard to convey to me?

s75
00:07:00.160 --> 00:07:03.360
And the immediate response to that is that yeah, we have browser agents.

s76
00:07:03.360 --> 00:07:04.800
We have computer user agents.

s77
00:07:04.800 --> 00:07:09.440
have very smart agents or assistants that can browse the web for us.

s78
00:07:09.440 --> 00:07:13.840
But that doesn't make sense because that that's that's a faster horses solution.

s79
00:07:13.840 --> 00:07:27.520
Why would I let an agent that doesn't know anything about the human perception limitations work with uh filters and pagenations and sorting and all of those UIs that we spent decades perfecting for us

s80
00:07:27.520 --> 00:07:29.759
um just for it to fulfill its task.

s81
00:07:29.759 --> 00:07:39.520
So Google has a um Google is very bullish on the other end of the spectrum which is co- browsing and and websites and they have a protocol called or standard called WebMCP.

s82
00:07:39.520 --> 00:07:51.759
WebMCP says okay instead of the agent taking screenshots of the websites and trying to figure out what's happening let's have the website expose some tools in the website JavaScript tools and the agent will walk with it.

s83
00:07:51.759 --> 00:08:02.160
And you can see here an example of Gemini in Chrome that um buys stuff for me or uh finds uh deals for me as I browse the web, which is very nice, but it's very simple.

s84
00:08:02.160 --> 00:08:07.360
What happens if we push Salesforce dashboard to Gemini on Chrome?

s85
00:08:07.360 --> 00:08:08.319
Why would I want that?

s86
00:08:08.319 --> 00:08:16.080
Why would I want to browse to Salesforce dashboard with Gemini to click buttons for me in a dashboard that I don't even want to use?

s87
00:08:16.080 --> 00:08:20.479
I mean, this is definitely not what we want.

s88
00:08:20.479 --> 00:08:26.720
And the shift is that assistants are becoming our entry point to the web, right?

s89
00:08:26.720 --> 00:08:28.240
We see it everywhere.

s90
00:08:28.240 --> 00:08:32.399
Every major lab wants its app to be the everything app.

s91
00:08:32.399 --> 00:08:34.399
And the shift is happening.

s92
00:08:34.399 --> 00:08:41.120
And friends of mine that connect Jira's MCP to their IDs, they don't go to Jira's website anymore.

s93
00:08:41.120 --> 00:08:41.680
Right?

s94
00:08:41.680 --> 00:08:43.599
And my mother uses CHP for everything.

s95
00:08:43.599 --> 00:08:48.880
If she could book a doctor's appointment using Chip, she won't go to that clinic's website anymore.

s96
00:08:48.880 --> 00:08:55.519
and you have a long tail of websites that no one is going to see because it's much easier for me to connect to them using my personal assistant.

s97
00:08:55.519 --> 00:09:01.760
So websites will become obsolete, browsers will become obsolete, and personal assistant will become our only gate to the web.

s98
00:09:01.760 --> 00:09:03.519
And I know what it sounds like.

s99
00:09:03.519 --> 00:09:07.680
It sounds like old men yelling at a lit literal cloud, right?

s100
00:09:07.680 --> 00:09:14.160
I mean, um, yeah, they're not going to go completely obsolete, but we see it with the younger generation today.

s101
00:09:14.160 --> 00:09:17.279
My 9-year-old goes to Chpt.

s102
00:09:17.279 --> 00:09:20.160
It doesn't go to Google when it when it wants to search for stuff.

s103
00:09:20.160 --> 00:09:30.720
I was uh I was traveling to Georgia, the country in Eastern Europe, and there was an older lady there that asked me to take a picture of her, like 70 year old, and on her phone, she had three apps,

s104
00:09:30.720 --> 00:09:32.959
WhatsApp, camera, and Chip.

s105
00:09:32.959 --> 00:09:33.680
That's it.

s106
00:09:33.680 --> 00:09:35.279
The shift is happening.

s107
00:09:35.279 --> 00:09:46.480
Um, and very smart people like David Ker from Sentry said a year ago, I will take this bet against anyone who thinks websites are going to be obsolete in 25 years.

s108
00:09:46.480 --> 00:09:48.640
That was one year ago.

s109
00:09:48.640 --> 00:09:58.399
Few days ago, David Kramer published designing for agents which says we have to recognize that interaction with products at century will no longer be exclusively through our web application

s110
00:09:58.399 --> 00:10:02.560
and we have to design for API first as a surface.

s111
00:10:02.560 --> 00:10:04.080
So everything is going headless.

s112
00:10:04.080 --> 00:10:12.000
Salesforce recently went headless which is very important because Salesforce main differentiator from its competitors is its UI.

s113
00:10:12.000 --> 00:10:14.079
It's the way that it conveys everything to the user.

s114
00:10:14.079 --> 00:10:14.959
But it went headless.

s115
00:10:14.959 --> 00:10:16.079
Cloudflow went headless.

s116
00:10:16.079 --> 00:10:17.440
Sentry went headless.

s117
00:10:17.440 --> 00:10:25.040
Um this is a a tweet by Cloudfl CEO saying that agents traffic surpassed uh human traffic in the web.

s118
00:10:25.040 --> 00:10:31.120
And we have this spectrum of interaction because if I if my agent is autonomous, I don't need this last mile of UI, right?

s119
00:10:31.120 --> 00:10:35.040
If I have an open claw, I just send and I just say yeah organize my emails.

s120
00:10:35.040 --> 00:10:37.040
It comes back said good.

s121
00:10:37.040 --> 00:10:41.600
But if I want to book a hotel for my honeymoon, I probably need this less mile of interaction.

s122
00:10:41.600 --> 00:10:44.880
So we call it the nearly headless web, right?

s123
00:10:44.880 --> 00:10:46.480
It's not completely headless.

s124
00:10:46.480 --> 00:10:55.200
Um the agents will interact with your web head with your um website headlessly, but then it will bring back these UI resources for you.

s125
00:10:55.200 --> 00:10:59.200
So user experience as we know it is being split.

s126
00:10:59.200 --> 00:11:06.399
Now it's the agent experience of your website plus how a user experiences the agent that experiences your website.

s127
00:11:06.399 --> 00:11:15.040
But 90% is going to rely on agent experiences which means your website needs to be agent ready for everything.

s128
00:11:15.040 --> 00:11:17.440
It needs to be agent ready for your agent.

s129
00:11:17.440 --> 00:11:20.720
If I'm booking I need to make bookings data ready for bookings agent.

s130
00:11:20.720 --> 00:11:31.839
I need to make it ready for my customers chpt or my customers open claw and I need I need to be ready for my human customer who doesn't have an agent but still wants to browse to booking.com

s131
00:11:31.839 --> 00:11:44.959
right so I need to to be agent ready and what we find found out is that agent ready means a lot of things so we heard a lot about AEO SEO geo how to get discovered but discover discovery is only the first

s132
00:11:44.959 --> 00:11:53.839
the first step because once an agent knows about you it still needs to know what you are and how to interact with you and how authenticate to you and how to headlessly pay you, right?

s133
00:11:53.839 --> 00:12:02.240
And actually when when we built uh it's an anecdote, but when we built analytics for our product, we asked Cloud Code, what's the best analytic service?

s134
00:12:02.240 --> 00:12:05.279
And it recommended Postto, which is an analytic service.

s135
00:12:05.279 --> 00:12:07.920
And we said no, we prefer Mixpanel because we know Mixpanel.

s136
00:12:07.920 --> 00:12:10.079
We worked with Mixpanel for a decade, we know how to work with it.

s137
00:12:10.079 --> 00:12:17.360
And Cloud Code insisted on Postto because it said Postto has better MCP and API and I can integrate to it better.

s138
00:12:17.360 --> 00:12:18.720
So we don't have brand loyalty, right?

s139
00:12:18.720 --> 00:12:31.680
We went with postto hog if that's what cloud code recommend but it made us think about mixpan mixpan spent a decade perfecting their UIX and developer experience we just left it just because cloud code prefers posto and it will always happen her for example

s140
00:12:31.680 --> 00:12:39.200
if it uses your product and it needs to spin up a browser to use your product it will remember that and next time it won't go to your product anymore

s141
00:12:39.200 --> 00:12:52.240
right so at aura which is an agentic web research lab we started researching uh we raised some raised some capital and we started researching what does it mean for the web to be ready for agents

s142
00:12:52.240 --> 00:13:01.440
and we have these um readiness benchmarks which is interesting so you can run it's free you can go to aura.ai AI you can run any website you get this

s143
00:13:01.440 --> 00:13:07.600
score and benchmarks according to a lot of protocols and best practices and you have agentic feedback.

s144
00:13:07.600 --> 00:13:24.320
So the agent actually returns the feedback about your uh your website and we have this leaderboard of how companies and products rank according to this benchmark and we started mapping the web and everything was nice.

s145
00:13:24.320 --> 00:13:31.839
But we hit one insight or one uh unexpected result.

s146
00:13:31.839 --> 00:13:40.240
We found out anyone here heard about lm.txt txt lm.txt that's like the de facto standard to be agent ready.

s147
00:13:40.240 --> 00:13:54.320
You say yeah if you if your website publish an lm.txt agent know how to interact with you um and you have o.md and pricing.mmd and x42 and a lot of standards and we found out that almost 50% of the website that we tested that we ran

s148
00:13:54.320 --> 00:14:02.480
published lms.txt but none of the agents that we ran on this website actually used llm.txt txt.

s149
00:14:02.480 --> 00:14:08.639
Actually, almost all the agents went straight to the docs page and then the homepage.

s150
00:14:08.639 --> 00:14:16.480
And the 40% that did use lm.txt use it only because the docs pointed out that there's a file called nmst txt that they need to to use.

s151
00:14:16.480 --> 00:14:17.920
So then it hit us.

s152
00:14:17.920 --> 00:14:24.800
We said it doesn't make a lot of sense for us as humans to define to agents what they need.

s153
00:14:24.800 --> 00:14:26.240
No one is doing it anymore.

s154
00:14:26.240 --> 00:14:35.120
Even OpenAI, they don't publish best practices for tools anymore because they say that every time they publish best practices, the models become better and these practices become obsolete.

s155
00:14:35.120 --> 00:14:44.480
Six months ago, the best practice for an MCP server was yeah uh have three paragraph descriptions so agents will know how to interact with you and now three lines are enough.

s156
00:14:44.480 --> 00:14:48.880
So best practices become we need the agents to define what the agent need, right?

s157
00:14:48.880 --> 00:14:51.600
We need agents feedback on this website.

s158
00:14:51.600 --> 00:14:58.480
So we built Aura's journey and Aura's journey that's a really cool that's also free.

s159
00:14:58.480 --> 00:15:09.600
You can go to journey.ora.ai and you can run on any website any intent with any agent and see the path of the agent as it tries to interact with the website.

s160
00:15:09.600 --> 00:15:22.240
So for example atio.com choose cl codec code and we run it on the website and you can see in real time how the agent goes and what it tries to look for in the website

s161
00:15:22.240 --> 00:15:33.120
and we we did it tens of thousands of times just to understand what agents really look for when they look when when they try to interact with with websites.

s162
00:15:33.120 --> 00:15:39.040
And the cool thing is not running just one harness is running multiple harnesses.

s163
00:15:39.040 --> 00:15:45.519
You see here cloud code and Eve which is Versel's harness um and CH GPT on the same website with the same intent.

s164
00:15:45.519 --> 00:15:45.760
Okay.

s165
00:15:45.760 --> 00:15:46.639
So this is cloud code.

s166
00:15:46.639 --> 00:15:48.800
This is haiku.

s167
00:15:48.800 --> 00:15:49.680
This is Eve.

s168
00:15:49.680 --> 00:15:52.399
See how different the agent journey looks like.

s169
00:15:52.399 --> 00:15:55.120
And this is CHP.

s170
00:15:55.120 --> 00:15:58.639
So CHP could find results better.

s171
00:15:58.639 --> 00:16:02.240
See you can just see the the the journey across across the website.

s172
00:16:02.240 --> 00:16:03.600
And we need to understand why.

s173
00:16:03.600 --> 00:16:06.320
We need to understand why does it happen?

s174
00:16:06.320 --> 00:16:13.199
Why websites publish o.md files but agent don't look for o.md files and what do they look for?

s175
00:16:13.199 --> 00:16:19.680
So in aura um you can actually uh go to any business business business question.

s176
00:16:19.680 --> 00:16:26.480
So for any domain you have like business goals or or questions and you can see the paths that the agents are taking.

s177
00:16:26.480 --> 00:16:33.120
And the uh interesting part of it is that okay now we know how agents interact with website.

s178
00:16:33.120 --> 00:16:39.199
What's the next big big milestone for the agentic web and for those of you who heard the the previous talk?

s179
00:16:39.199 --> 00:16:42.560
Um it's discovery, right?

s180
00:16:42.560 --> 00:16:44.560
But it's not the discovery that we think of.

s181
00:16:44.560 --> 00:16:53.120
It's not SEO, it's not GEO, it's not AEO because we have to remember that every revolution came with its discovery layer, right?

s182
00:16:53.120 --> 00:16:59.759
The web revolution came with search, um mobile came with app stores, social came with feeds.

s183
00:16:59.759 --> 00:17:02.000
What is the discovery layer for aic resources?

s184
00:17:02.000 --> 00:17:03.519
What is the discovery layer for MCPS?

s185
00:17:03.519 --> 00:17:06.480
What is the discovery layer for open API.json?

s186
00:17:06.480 --> 00:17:07.839
What do we even look for?

s187
00:17:07.839 --> 00:17:13.439
I mean, maybe Airbnb has a better MCP than than Booking.com, but Booking.com has better API than Airbnb.

s188
00:17:13.439 --> 00:17:15.600
So, how do we do it?

s189
00:17:15.600 --> 00:17:27.199
Um, web search, classic web search, it's not enough because um it it's based on human SEO and human uh page rank and it's not flexible enough.

s190
00:17:27.199 --> 00:17:38.160
custom registries like per agents or per chats, they're not enough because they're closed and they're uh um and and they require every app to to submit itself to these reg registries

s191
00:17:38.160 --> 00:17:41.200
and a central registry like an MCP registry.

s192
00:17:41.200 --> 00:17:43.280
That's not enough because who will do the curation?

s193
00:17:43.280 --> 00:17:44.400
What's the governance model?

s194
00:17:44.400 --> 00:17:48.559
How do we decide which resource comes to that registry?

s195
00:17:48.559 --> 00:17:50.880
There are emerging standards around it.

s196
00:17:50.880 --> 00:18:03.120
There's AI catalog.json JSON which is a standard by entropic openai, Google, MCP and A2A which standardizes how a website exposes itself to agents and there's the agentic resource discovery standard

s197
00:18:03.120 --> 00:18:13.120
which is by all these companies and more which is basically standardized how uh a discovery layer or or a directory uh expose itself to agents.

s198
00:18:13.120 --> 00:18:17.919
So we built or directory right because we're a research lab for the agentic web.

s199
00:18:17.919 --> 00:18:27.679
We built this and in or directory we take all the domains that we scanned or scanned the results and we put it in a directory that agents can actually query right

s200
00:18:27.679 --> 00:18:30.960
um we expose the AI catalog.json JSON files.

s201
00:18:30.960 --> 00:18:42.559
So you can see here for example for monday.com you can see that we generate this JSON file that basically tells agents yeah this is the MCP server for Monday this is the API server from monday.com

s202
00:18:42.559 --> 00:18:53.600
and you can just the agent can just query that right so for every domain it can just goes to go to aic catalog.json JSON and we expose the registry the directory itself.

s203
00:18:53.600 --> 00:19:05.760
So for every entry on the directory we can uh we can tell the agent yeah for example versel these are the agentic resources um and this is and this is how you access them and obviously

s204
00:19:05.760 --> 00:19:09.760
this directory is fully ard or aentic resource discovery compliant.

s205
00:19:09.760 --> 00:19:13.520
So any agent can just query that directory with any with any query.

s206
00:19:13.520 --> 00:19:15.039
Right?

s207
00:19:15.039 --> 00:19:16.799
So this is uh this is aura.

s208
00:19:16.799 --> 00:19:23.440
directory uh if you want to take a look all of it is part of the aura multiverse journey directory and the ranker.

s209
00:19:23.440 --> 00:19:26.080
And we found out some very other interesting insights.

s210
00:19:26.080 --> 00:19:38.320
For example being agent ready and being human accessible is very similar because LLMs when they come to your website they're like um users or people with vision disabilities.

s211
00:19:38.320 --> 00:19:39.760
they don't see your website.

s212
00:19:39.760 --> 00:19:42.480
They need other signals to understand how to walk with your website.

s213
00:19:42.480 --> 00:19:48.000
And making your website human accessible helps agent accessibility and vice vice versa.

s214
00:19:48.000 --> 00:19:51.120
So to wrap up, these are MCP apps.

s215
00:19:51.120 --> 00:19:54.080
This is what I've been working on in the past few months.

s216
00:19:54.080 --> 00:19:57.120
And this was the last piece for the agentic web.

s217
00:19:57.120 --> 00:19:59.200
This actually brought the agentic web.

s218
00:19:59.200 --> 00:20:02.160
And now agents start to roam the web.

s219
00:20:02.160 --> 00:20:03.440
And agents need different things.

s220
00:20:03.440 --> 00:20:04.480
They need different roads.

s221
00:20:04.480 --> 00:20:06.160
They need different infras.

s222
00:20:06.160 --> 00:20:08.799
But we don't want to rebuild the web for agents.

s223
00:20:08.799 --> 00:20:11.600
We we want to make the web agent accessible.

s224
00:20:11.600 --> 00:20:17.840
So let's just make sure the web is prepared for when agents are coming.

s225
00:20:19.120 --> 00:20:21.679
Thank you very much.

s226
00:20:22.244 --> 00:20:24.244
[applause]