WEBVTT

NOTE Sentence-level transcript of https://www.youtube.com/watch?v=-jY2T2PiJBE

NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/-jY2T2PiJBE.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
00:00:12.640 --> 00:00:13.440
Hi.

s3
00:00:13.440 --> 00:00:15.400
So, hi everyone.

s4
00:00:15.400 --> 00:00:18.720
We built this talk yesterday, so it might be out of date.

s5
00:00:18.720 --> 00:00:26.000
I'm Ido Sadan, I am the creator of MCPY and co-creator and maintainer of MCP apps in the MCP steering committee.

s6
00:00:26.000 --> 00:00:30.200
I also created Adam Craft if you were in the talk yesterday.

s7
00:00:30.200 --> 00:00:30.840
I'm the Adi.

s8
00:00:30.840 --> 00:00:32.119
I work with Ido on MCPY.

s9
00:00:32.119 --> 00:00:39.880
I'm also the co-creator and maintainer of the MCP apps spec and recently co-founded Aura, which is a research lab for the agentic web.

s10
00:00:39.880 --> 00:00:44.120
And we're going to talk a little bit more about it later.

s11
00:00:45.680 --> 00:00:47.840
So, MCP apps are all around us.

s12
00:00:47.840 --> 00:00:59.000
You might not even realize it, but all the fancy apps you have today in ChatGPT and VS Code and Slack are actually all based on MCP and the MCP app spec.

s13
00:01:01.040 --> 00:01:05.000
And if we take a step back and we ask, why do we need MCP apps?

s14
00:01:05.000 --> 00:01:07.600
What's the idea behind MCPY or MCP apps?

s15
00:01:07.600 --> 00:01:18.560
So, when we work with chats, when we chat client clients, we used to text because that's the natural interface, but text is really the worst way to convey a lot of information, right?

s16
00:01:18.560 --> 00:01:20.720
Because we don't want walls of text.

s17
00:01:20.720 --> 00:01:25.280
And actually, this is the main blocker from companies to build an MCP server.

s18
00:01:25.280 --> 00:01:27.840
They don't want to be reduced to a textual database.

s19
00:01:27.840 --> 00:01:30.440
They don't want to lose their brand identity in the process.

s20
00:01:30.440 --> 00:01:37.880
They don't want their data that they work so hard on um building the UX for to look something like this.

s21
00:01:37.880 --> 00:01:46.920
So, instead of this, what if the apps could just send their UI to the chat, right?

s22
00:01:46.920 --> 00:01:51.600
What if every service and every brand could just send their user interface to the chat?

s23
00:01:51.600 --> 00:02:03.760
So, instead of us looking at something like this, we could just have the apps send their own identity, their own UI chunks into the chat, and then we take a look and we see, "Okay, yeah, I know this is Shopify in the middle.

s24
00:02:03.760 --> 00:02:05.080
I know this is Hugging Face.

s25
00:02:05.080 --> 00:02:06.360
I know this is Monday."

s26
00:02:06.360 --> 00:02:10.640
And what if we don't want to do it only as a visualization?

s27
00:02:10.640 --> 00:02:12.680
We also want to do it interactive.

s28
00:02:12.680 --> 00:02:18.320
So, we want the users to be able to actually interact with Hugging Face, for example.

s29
00:02:18.360 --> 00:02:22.520
And for Hugging Face to actually do something with it.

s30
00:02:22.600 --> 00:02:35.360
So, we don't have to imagine the future as we said with MCPUI, which I created in May last year, and took that, which is essentially like an open protocol for

s31
00:02:35.360 --> 00:02:37.959
interactive applications over MCP.

s32
00:02:37.959 --> 00:02:46.880
So, it's not only how you transmit UI, but also how that UI, that application connect connect communicates with the host.

s33
00:02:46.880 --> 00:03:00.200
And just a few months ago, we partnered with Anthropic and OpenAI to create the official extension to MCP, which we call MCP apps based on MCPUI, MCP SDK, and other

s34
00:03:00.200 --> 00:03:01.880
solutions in the field.

s35
00:03:01.880 --> 00:03:13.040
Their launch was pretty cool with Claude and VS Code supporting it to begin with, but now obviously also OpenAI and others have adopted it.

s36
00:03:13.239 --> 00:03:17.320
Yeah, and there are a lot of early adopters to MCPUI.

s37
00:03:17.320 --> 00:03:19.760
11 Labs, Shopify, Postman.

s38
00:03:19.760 --> 00:03:23.880
Those were one of the first first companies to support it back like a year ago.

s39
00:03:23.880 --> 00:03:27.519
They were the one believing in this spec, in this vision.

s40
00:03:27.519 --> 00:03:30.040
And Goose also supported it.

s41
00:03:30.040 --> 00:03:36.519
And it's a it's a funny anecdote because today Block released their agentic commerce solution that is based on MCP apps.

s42
00:03:36.519 --> 00:03:45.000
So, a year ago Goose was the first client to support MCPUI, and now it is part of Block's product uh product.

s43
00:03:45.000 --> 00:03:49.720
And today we have a lot more clients that are supporting MCPUI.

s44
00:03:49.720 --> 00:03:57.840
We We Cursor, and we have co-pilot and GitHub ChatGPT support MCP apps.

s45
00:03:57.840 --> 00:04:08.040
ChatGPT apps that you know are actually based on MCP apps and open eye actually recommend using MCP apps as the protocol to build ChatGPT apps.

s46
00:04:08.040 --> 00:04:12.480
Postman and a lot more and obviously Cloud supports MCP apps.

s47
00:04:12.480 --> 00:04:15.240
But we also have a lot large community around it, right?

s48
00:04:15.240 --> 00:04:24.040
So people start to to build plugins to MCP apps and um integrations to different agents and also courses on how to build MCP apps.

s49
00:04:24.040 --> 00:04:26.440
This is by integration for MCP apps.

s50
00:04:26.440 --> 00:04:28.840
So we have a lot community around it.

s51
00:04:28.840 --> 00:04:41.520
Um There's a repo X app which is the repo for MCP apps where everyone can just come and propose PRs and ideas of how to how to extend this spec and we have a work group in the MCP committee

s52
00:04:41.520 --> 00:04:43.080
and we're convening every 3 weeks.

s53
00:04:43.080 --> 00:04:51.640
We have a tri-weekly meeting on the future of the protocol and how to make the spec not just serve the bigger apps but also the community.

s54
00:04:51.640 --> 00:04:58.560
So it's an open working group with Anthropic, Open AI and all the partners in in the MCP apps protocol.

s55
00:04:58.680 --> 00:05:03.320
Okay, so let's look at a few of the core concepts of MCP apps.

s56
00:05:03.320 --> 00:05:08.720
The first and most obvious one is how do we even transmit UI over MCP?

s57
00:05:08.720 --> 00:05:20.760
So if we look at this example of Cloud like you know agent times like a few months ago and I would ask something best case scenario it would reach out to my MCP server

s58
00:05:20.760 --> 00:05:26.520
and it would get back a textual response which is obviously suboptimal.

s59
00:05:26.520 --> 00:05:31.040
So let's say I do want to get some something better.

s60
00:05:31.040 --> 00:05:38.280
So now I can use existing MCP primitives like a resource and now return HTML.

s61
00:05:38.280 --> 00:05:49.960
And I can take that HTML and since Cloud supports MCP apps it can turn it into an interactive application of the best soundtrack in the world.

s62
00:05:50.520 --> 00:05:53.800
And what if you wanted to be really interactive, right?

s63
00:05:53.800 --> 00:05:55.880
This is nice because it shows the best soundtrack in the world.

s64
00:05:55.880 --> 00:05:58.600
What if I want to favorite one of the songs there?

s65
00:05:58.600 --> 00:06:00.600
I want interaction.

s66
00:06:00.600 --> 00:06:04.760
I want communication between the app and the host.

s67
00:06:04.760 --> 00:06:09.880
So, when the user clicks on the favorite button, MCP apps actually standardizes this flow.

s68
00:06:09.880 --> 00:06:19.360
So, instead of the app sending a message to to the backend, to Spotify's backend, it's actually sending a message to the host saying, "Hey, user clicked a button.

s69
00:06:19.360 --> 00:06:20.400
Do something with it.

s70
00:06:20.400 --> 00:06:23.760
I recommend you to call a tool in Spotify's MCP server."

s71
00:06:23.760 --> 00:06:25.040
And the host decides what to do.

s72
00:06:25.040 --> 00:06:27.240
The host keeps this control of the flow.

s73
00:06:27.240 --> 00:06:31.040
In this case, the host can decide to actually call the favorite favorite tool.

s74
00:06:31.040 --> 00:06:35.040
And MCP apps standardizes this flow.

s75
00:06:35.040 --> 00:06:36.640
Okay, so seeing is believing.

s76
00:06:36.640 --> 00:06:40.000
So, let's see an example from Claude.

s77
00:06:42.440 --> 00:06:43.200
Yeah.

s78
00:06:43.200 --> 00:06:48.320
Uh so, let's say that I'm a product manager to understand the status of my funnel.

s79
00:06:48.320 --> 00:06:51.680
So, I would go to Claude and I would ask what's the status?

s80
00:06:51.680 --> 00:06:56.160
In the again, old world of a few months ago, uh I would get back the textual response.

s81
00:06:56.160 --> 00:06:58.240
Let's say that it's PostHog.

s82
00:06:58.240 --> 00:07:01.520
So, it reached out to the PostHog server, got back the textual response.

s83
00:07:01.520 --> 00:07:03.919
It's factually correct, but it's useless.

s84
00:07:03.919 --> 00:07:08.040
I mean, how do I even take that and understand quickly what's going on?

s85
00:07:08.040 --> 00:07:10.560
I would have to read, which I don't want to do.

s86
00:07:10.560 --> 00:07:12.680
Uh and it's pretty challenging.

s87
00:07:12.680 --> 00:07:21.760
Uh but luckily, because both PostHog server and Claude as a host support MCP apps, I can just say, "Show me."

s88
00:07:21.760 --> 00:07:32.280
And now, instead of getting that block of text, I can actually get something useful, uh which is this interactive um widget that you would get, you know, on the PostHog

s89
00:07:32.280 --> 00:07:33.440
uh uh server.

s90
00:07:33.440 --> 00:07:36.640
And when you have that, you can at a glance see what's going on.

s91
00:07:36.640 --> 00:07:39.000
And as you can see, it's branded PostHog.

s92
00:07:39.000 --> 00:07:46.960
So, you're actually getting the PostHog experience within ChatGPT or Claude, etc. Uh but it doesn't really end there.

s93
00:07:46.960 --> 00:07:50.520
As we said, MCP apps is also like an interactive photo call.

s94
00:07:50.520 --> 00:07:58.000
So, not only can I see and and interact with it, I can also do stuff like ask him to explain what a funnel is.

s95
00:07:58.000 --> 00:07:59.400
I might not even know that.

s96
00:07:59.400 --> 00:08:09.200
So, again, instead of getting that huge wall of text explaining what a funnel is, I can just get this generative UI answer from Claude, which uses MCP apps.

s97
00:08:09.200 --> 00:08:15.840
It streams like the HTML inside, and now I can get this nice interactive experience of learning.

s98
00:08:15.840 --> 00:08:22.520
And not only is it visually nice and helps me understand, but it's also fully interactive.

s99
00:08:22.520 --> 00:08:29.080
And when we say interactive, it actually means that clicking it would help me communicate with the host.

s100
00:08:29.080 --> 00:08:33.320
So, let's say that I want to understand like a particular step in the funnel.

s101
00:08:33.320 --> 00:08:43.960
Uh I just go and I click on it, and since it's an MCP app, it can send a prompt back to the uh model and say, "Okay, explain this specific step to me."

s102
00:08:43.960 --> 00:08:46.680
And I can advance the flow.

s103
00:08:46.680 --> 00:08:51.240
Uh so, this is a like an example of of how that uh looks.

s104
00:08:51.240 --> 00:08:52.440
So, how does it actually work?

s105
00:08:52.440 --> 00:08:57.120
If you look at the architecture of it, uh so, we started by prompting.

s106
00:08:57.120 --> 00:08:58.600
So, we type something in.

s107
00:08:58.600 --> 00:09:01.000
Uh we asked for the funnel information.

s108
00:09:01.000 --> 00:09:02.880
A tool call went out.

s109
00:09:02.880 --> 00:09:08.960
Since our server supports MCP apps, that tool call is actually linked to a resource.

s110
00:09:08.960 --> 00:09:15.640
And if you look at the uh code here, then, you know, it's it's a it's just a resource with a uh some prefix.

s111
00:09:15.640 --> 00:09:16.520
Uh we take that.

s112
00:09:16.520 --> 00:09:17.920
It's pretty simple code.

s113
00:09:17.920 --> 00:09:21.640
I could just add the but it's still the the resource with the HTML, and you're done.

s114
00:09:21.640 --> 00:09:27.040
Uh that resource is then um consumed by the host.

s115
00:09:27.040 --> 00:09:30.560
In practice, it's usually consumed beforehand, like it's preloaded.

s116
00:09:30.560 --> 00:09:33.360
Uh but imagine that it's just consumed in real time.

s117
00:09:33.360 --> 00:09:37.840
That same HTML then passed to the host that also supports MCP apps.

s118
00:09:37.840 --> 00:09:51.600
MCP apps basically if you look at the MCP UI SDK, just a React component or a web component that just accepts that resource plus a callback which is how we implement that communication protocol as I said earlier.

s119
00:09:51.600 --> 00:09:54.000
And renders it in a sandbox.

s120
00:09:54.000 --> 00:09:57.800
So, like we said, not only is it presentational, I can click.

s121
00:09:57.800 --> 00:09:59.440
So, what happens when I click?

s122
00:09:59.440 --> 00:10:04.120
So, we click on it, it sends back through that callback the event all the way up.

s123
00:10:04.120 --> 00:10:14.000
The model takes that event and then it can send out a tool call or call a resource or anything else that's completing the agentic flow.

s124
00:10:14.280 --> 00:10:18.200
And this architecture actually brings a new philosophy or a new vision to the web.

s125
00:10:18.200 --> 00:10:30.520
So, instead of us thinking of the web as tabs or services that we need to consume using a browser, we're now consuming it using our own personal assistants, right?

s126
00:10:30.520 --> 00:10:31.360
What does it mean?

s127
00:10:31.360 --> 00:10:36.600
It means that if I want to accomplish a task, for example, plan a um anniversary.

s128
00:10:36.600 --> 00:10:43.320
So, up until now I had to open 20 tabs in the browser and I had to try to convey my intent to each of those services.

s129
00:10:43.320 --> 00:10:49.760
And by saying conveying my intent, it means that I have to interact with the dashboards or the UIs of those companies.

s130
00:10:49.760 --> 00:11:00.839
So, just to plan an anniversary, I need to convey my intent to Google Calendar and Amazon and Booking and Booking again and Amazon again and all and I don't need 99%

s131
00:11:00.839 --> 00:11:03.520
of the UI that is shown there because this UI doesn't know me.

s132
00:11:03.520 --> 00:11:05.280
It doesn't have the context on me.

s133
00:11:05.280 --> 00:11:10.360
What if we could just take these UIs and just break them into atoms?

s134
00:11:10.360 --> 00:11:14.080
And those atoms can be composed by my own personal assistant, right?

s135
00:11:14.080 --> 00:11:16.160
Because I don't need the the UI.

s136
00:11:16.160 --> 00:11:17.720
I need those atoms.

s137
00:11:17.720 --> 00:11:27.960
So, if we can take these atoms and have my cloud or ChatGPT or OpenCloud just use them using MCP UI, we can have this flow.

s138
00:11:27.960 --> 00:11:30.760
So, my proactive assistant can say, "Yeah, I know.

s139
00:11:30.760 --> 00:11:37.280
I see that you have an anniversary coming and instead of just showing me data from Google Calendar, it can display a Google Calendar chunk.

s140
00:11:37.280 --> 00:11:40.760
Now, for me it's good because I know Google Calendar, I trust Google.

s141
00:11:40.760 --> 00:11:48.520
For Google it's good because it maintains their brand and identity and for the host it's good because they don't need to develop these capability themselves.

s142
00:11:48.520 --> 00:11:57.680
And it goes even deeper because if I'm interacting with Amazon, instead of Amazon being reduced to just a list of items or or text, I can see Amazon.

s143
00:11:57.680 --> 00:12:05.440
I can I can know that this is this is Amazon and I can complete my entire flow without even leaving my assistant.

s144
00:12:05.440 --> 00:12:07.480
And this is the agentic web.

s145
00:12:07.480 --> 00:12:12.600
This is how we're going to consume the web because my assistant will have the context on me.

s146
00:12:12.600 --> 00:12:15.320
It It will know to pull the the map from booking.com.

s147
00:12:15.320 --> 00:12:17.360
I don't need to know that, right?

s148
00:12:17.360 --> 00:12:25.760
So, this is going to be the shift that we're going to see very soon where websites are going to shift into small chunks of UIs inside inside personal assistants.

s149
00:12:25.760 --> 00:12:26.053
Um

s150
00:12:26.053 --> 00:12:26.280
[snorts]

s151
00:12:26.280 --> 00:12:35.880
and with that come new interaction mindset because um if I click on something in the Shopify's MCP app, then Shopify doesn't control my journey anymore.

s152
00:12:35.880 --> 00:12:37.160
The host does.

s153
00:12:37.160 --> 00:12:39.680
Um and no application will control the user journey anymore.

s154
00:12:39.680 --> 00:12:43.080
So, Amazon won't be able to know to see my flow.

s155
00:12:43.080 --> 00:12:46.440
It everything will go through the chat for auditability.

s156
00:12:46.440 --> 00:12:53.640
Um and MCP apps actually standardizes it by defining this three level of control over the user journey.

s157
00:12:53.640 --> 00:13:02.480
So, an app can notify the chat that something happened or an app can actually ask the chat to run a prompt and and releasing all responsibility to the chat.

s158
00:13:02.480 --> 00:13:13.520
So, MCP apps actually standardizes it and this is the new software flow, the new flow of interaction that we're going to see between applications, the chats, and the users.

s159
00:13:13.520 --> 00:13:23.440
Um in 2026, we had we had an amazing year of standardizing MCP UI and 2026 is going to be the year where it's going to be a global standard for UI.

s160
00:13:23.440 --> 00:13:23.680
Yeah.

s161
00:13:23.680 --> 00:13:25.240
But, it's still evolving.

s162
00:13:25.240 --> 00:13:27.120
There's a lot of stuff going on.

s163
00:13:27.120 --> 00:13:34.520
Even in these past few months, these are some of the things that are already in or already contributed or proposed uh, by the community.

s164
00:13:34.520 --> 00:13:39.640
Uh, so you still have a lot of time and a lot of room to influence how this future will look like.

s165
00:13:39.640 --> 00:13:41.680
Uh, so you can go to X apps.

s166
00:13:41.680 --> 00:13:45.200
Uh, that's the official SDK and spec is also hosted there.

s167
00:13:45.200 --> 00:13:48.480
It's under the official model context protocol uh, repository.

s168
00:13:48.480 --> 00:13:49.880
There will be a QR code later.

s169
00:13:49.880 --> 00:13:53.560
Uh, so you don't have to uh, to uh, photograph it.

s170
00:13:53.560 --> 00:14:04.680
Uh, and also um, the the cool thing about using X apps in particular is that because it's maintained by us directly, uh, all changes to the spec are immediately

s171
00:14:04.680 --> 00:14:06.040
uh, reflected in the SDK.

s172
00:14:06.040 --> 00:14:10.760
So, if you use that SDK, then you automatically get all the new stuff out of the back.

s173
00:14:10.760 --> 00:14:13.320
Uh, these are some of the issues that we have.

s174
00:14:13.320 --> 00:14:16.240
So, please feel free to come and uh, contribute.

s175
00:14:16.240 --> 00:14:17.400
So, what's next?

s176
00:14:17.400 --> 00:14:20.520
Um, there's a bunch of stuff coming up.

s177
00:14:20.520 --> 00:14:27.120
Uh, the first thing uh, that we get a lot of uh, uh, of asked for is kind of reusable views.

s178
00:14:27.120 --> 00:14:36.440
So, if you have uh, um, companies like Autodesk that have really heavy apps like they have in, you know, the entire 3D render there.

s179
00:14:36.440 --> 00:14:42.640
They don't want to keep re-rendering that over and over again because it just it takes time, it's inefficient.

s180
00:14:42.640 --> 00:14:46.520
Uh, that is the way that we had to do it uh, for the MVP.

s181
00:14:46.520 --> 00:14:56.520
But we are working on thinking of maybe we can pass some identifier from the server uh, in a way that would help the model actually keep updating the same view.

s182
00:14:56.520 --> 00:14:59.960
Uh, the other way to do this is

s183
00:14:59.960 --> 00:15:07.880
Um, app tools, which is something uh, if you've heard of web MCP, which is Google standard of how agents will interact in with web views.

s184
00:15:07.880 --> 00:15:10.760
So, in MCP us, we actually standardize it into app tools.

s185
00:15:10.760 --> 00:15:17.280
So, up until now we saw the flow where users does something in the app and the app talks to the host.

s186
00:15:17.280 --> 00:15:20.000
But what if the host or the chat wants to speak to the app?

s187
00:15:20.000 --> 00:15:25.800
If the user writes something, uh, fill out this form for me and the chat will fill out the form for the user.

s188
00:15:25.800 --> 00:15:29.720
So, MCP apps actually standardizes this this flow which we call view tools.

s189
00:15:29.720 --> 00:15:31.600
That's actually that's in the spec right now.

s190
00:15:31.600 --> 00:15:34.720
It's also it's going to be released very soon.

s191
00:15:34.720 --> 00:15:38.920
And we're working on this generative UI spectrum where you have predefined UI.

s192
00:15:38.920 --> 00:15:39.720
That's MCP apps.

s193
00:15:39.720 --> 00:15:45.560
That's like the black box iframe that renders all trace UI in that example.

s194
00:15:45.560 --> 00:15:51.080
But you also have other things on this spectrum like declarative UI like JSON render or A2UI.

s195
00:15:51.080 --> 00:15:58.040
These specs that say yeah, the the app just returns an instructions on how to build the UI, but the chat will actually build the UI.

s196
00:15:58.040 --> 00:16:01.120
And you have fully generative UI on the other end of the spectrum.

s197
00:16:01.120 --> 00:16:06.080
And if you know cloud apps, yeah, MCP apps is agnostic to the way the UI is generated.

s198
00:16:06.080 --> 00:16:12.040
And if you know cloud apps imagine feature where you can just ask cloud to generate a UI for you.

s199
00:16:12.040 --> 00:16:14.160
That's actually based on MCP apps.

s200
00:16:14.160 --> 00:16:18.480
So, this is an MCP app behind the scenes, but it supports generative UI.

s201
00:16:18.480 --> 00:16:22.360
So, we're working on interoperability with those other standards.

s202
00:16:22.360 --> 00:16:33.400
And actually just a few days ago we released a guide on how to do A2UI versus a generative UI standard and MCP apps which is the standard.

s203
00:16:33.400 --> 00:16:35.040
How to do interoperability.

s204
00:16:35.040 --> 00:16:42.880
How can a server can write A2UI and ship it to Gemini, but also wrap it as an MCP app to ship to ChatGPT and vice versa.

s205
00:16:42.880 --> 00:16:44.960
An MCP app is supported everywhere.

s206
00:16:44.960 --> 00:16:46.080
So, it can run everywhere.

s207
00:16:46.080 --> 00:16:52.000
If you build it once, it runs in Libra Chat which is an open source MCP app supported in ChatGPT.

s208
00:16:52.000 --> 00:16:58.080
That's the same app that you're seeing the same code base that runs in in both which is pretty cool.

s209
00:16:58.080 --> 00:17:00.080
Yeah.

s210
00:17:00.400 --> 00:17:01.560
Yeah.

s211
00:17:01.560 --> 00:17:04.839
So, this isn't just a technology or a cool feature.

s212
00:17:04.839 --> 00:17:08.160
This is an entirely new way to distribute applications.

s213
00:17:08.160 --> 00:17:20.120
So, if you look just a few months back then someone said that ChatGPT in particular has 800 million weekly users which is 10% of the entire world population.

s214
00:17:20.120 --> 00:17:21.920
That's insane.

s215
00:17:21.920 --> 00:17:27.640
So, if you think about the web in general, it took around 13 years to get to that number of users.

s216
00:17:27.640 --> 00:17:34.920
So, if you look at that and you think that in the last few months we actually had a growth of over 1 billion.

s217
00:17:34.920 --> 00:17:42.120
Just that we have like 170 times the total addressable market of the Apple App Store when it launched.

s218
00:17:42.120 --> 00:17:43.960
So, MCP apps are everywhere.

s219
00:17:43.960 --> 00:17:49.240
So, actually to list them, it is called cloud, open AI, etc. It's already there.

s220
00:17:49.240 --> 00:17:51.120
So, how do you get started?

s221
00:17:51.120 --> 00:17:55.960
You can clone those you can go to the X apps.

s222
00:17:55.960 --> 00:17:59.200
As a host also go to X apps or the MCP website.

s223
00:17:59.200 --> 00:18:01.417
Please visit the official repo.

s224
00:18:01.417 --> 00:18:01.880
[laughter]

s225
00:18:01.880 --> 00:18:04.400
The X apps repo to get involved.

s226
00:18:04.400 --> 00:18:05.520
And yeah.

s227
00:18:05.520 --> 00:18:06.920
So, embrace the new web.

s228
00:18:06.920 --> 00:18:08.200
It's awesome.

s229
00:18:08.200 --> 00:18:11.480
With MCP apps you can write once and run it everywhere.

s230
00:18:11.480 --> 00:18:13.560
And the future is looking bright.

s231
00:18:13.560 --> 00:18:17.040
Not quite Travis, but with MCP and MCP apps we're close.

s232
00:18:17.040 --> 00:18:18.880
And come talk to us afterwards.

s233
00:18:18.880 --> 00:18:21.320
Yeah, thank you.

s234
00:18:36.132 --> 00:18:38.132
[music]