MCP Apps Are Changing the Internet. Here’s How to Build One! (Skybridge)

BBetter Stack
Computing/SoftwareSmall Business/StartupsInternet Technology

Transcript

00:00:00There's a huge paradigm shift happening across the industry because of AI.
00:00:04More and more users are using chatbots nowadays for their routine searches or research,
00:00:10and that's why the concept of MCP apps is getting more and more popular.
00:00:14Instead of spitting out a JSON file or a markdown document,
00:00:18more apps and services are switching to providing these interactive and helpful widgets inside a chatbot.
00:00:24And the company Alpic recently released a cool new framework called SkyBridge
00:00:29that makes it super easy to create your own custom MCP apps for all sorts of different use cases.
00:00:35So in today's video, we'll take a look at SkyBridge, see how it works, and try it out for ourselves.
00:00:41It's going to be a lot of fun, so let's dive into it.
00:00:47To really understand why SkyBridge is such a big deal, you have to understand how these new apps work.
00:00:53When you build a traditional web app, you are building it for one user, the human.
00:00:58But with an MCP app, you're actually building for two users at once, the human and the AI assistant,
00:01:05because they both share the exact same interface.
00:01:08The human can click buttons, type inputs, and slide sliders, and the LLM instantly sees that state.
00:01:15And conversely, the LLM can update the data in the background,
00:01:18and the UI changes in real time right in front of you.
00:01:21And SkyBridge is an open source TypeScript framework for that that handles all the complex plumbing,
00:01:26the protocol bridging, the state sync, and the security rules,
00:01:30so you can just focus on writing standard React code.
00:01:34Now, if you've ever tried to build MCP apps for ChatGPT or Claude before,
00:01:39you know that the local testing loop used to be a massive headache.
00:01:43You had to manually set up public tunnels, copy-paste configurations,
00:01:47and restart your chat sessions just to fix a tiny CSS typo.
00:01:52But the coolest part of SkyBridge 1.0 is that it completely redesigned the DevTools control panel.
00:01:59Now, when you run your local dev server, it fires up an emulator dashboard in your browser.
00:02:04And inside this dashboard, you have three nice tools.
00:02:08The first one is the Alpik Playground,
00:02:11a local sandbox where you can test your tools and tweak your React widgets using live hot module replacement.
00:02:17And you can see the changes instantly without touching an actual LLM yet.
00:02:21And secondly, there's an integrated tunnel,
00:02:24which allows you with literally a single click to expose your local server through a secure public URL.
00:02:30So you don't have to mess around with external tunneling anymore.
00:02:33Just paste this URL into ChatGPT's developer settings and you're live.
00:02:37And third is the Beacon Audit tool.
00:02:40And this is a lifesaver if you plan to submit your apps to public stores.
00:02:44It scans your app's metadata, tools, and security policies to catch common rejection triggers
00:02:49before you submit your app for review.
00:02:52And the coolest thing is that even if you've never built an MCP app,
00:02:56you can easily install the SkyBridge skill onto your favorite AI agent to get started instantly.
00:03:02So now let's try to build a small MCP app on our own and see how it works.
00:03:07So for this demo, I've went ahead and created this little e-commerce website
00:03:11where you can shop for different camera lenses.
00:03:14And it's basically a fully built React-based store with a cart and checkout functionality.
00:03:19And now I want to add an MCP app for it.
00:03:22So that if a potential customer uses a chatbot like ChatGPT or Claude to search for camera lens recommendations,
00:03:30it will immediately call our MCP app to show our product listings for the end user.
00:03:36So as I said before, the easiest way to get started is just to install the SkyBridge skill.
00:03:42And we can install it globally or just for our specific repo.
00:03:46And for this demo purpose, I'm going to go ahead and install it in the repo.
00:03:51And now you can see we have this .agents directory with the SkyBridge skill present.
00:03:57And all I have to do now is just ask Claude to build me an MCP app for my e-commerce shop.
00:04:04And before building it, Claude will first ask me some follow-up questions about the intent of the app
00:04:09to better understand what kind of functionality we need to include in it.
00:04:13And for this demo, I want to have an app that lists lenses based on what the user is looking for.
00:04:18And then we can select them and compare them side by side and also add them to our cart
00:04:24and provide a quick checkout solution for our customers.
00:04:28And that's basically all I need to do.
00:04:29And now I just let Claude do its thing.
00:04:32And in about a few minutes, we see that Claude has generated our MCP app.
00:04:37And the coolest part is that now we can open the developer dashboard provided by SkyBridge
00:04:41and try it out for ourselves.
00:04:44So now if we open the dashboard, we can see how it looks.
00:04:47And here we have our search lenses tool and the place order tool.
00:04:53So first of all, for search lenses, I'm just going to insert the max price for the lenses I'm looking for.
00:05:00And let's limit it to five lenses maybe.
00:05:03And there you go.
00:05:04It found 14 matches that are under $2,000.
00:05:09Here's how the app looks on desktop.
00:05:12We can also preview the mobile view.
00:05:14Looks good.
00:05:15We also have a dark mode and a light mode.
00:05:18And here we have the tool output and the state.
00:05:23And watch what happens if I select some of the lenses.
00:05:26As you can see, the state updates on the fly, which is super cool.
00:05:31So now let's choose three lenses.
00:05:34And now if we click compare, we now get to the compare screen.
00:05:40And here we can see all our three lenses side by side with all the parameters.
00:05:45And if we're happy with this selection, we can also go to the checkout.
00:05:49And this moves to the checkout stage where we can put in our full name, email, and shipping address and place our order.
00:05:59So as you can see, this dashboard is really cool for testing out your widgets to see if everything is working as expected.
00:06:08And now if we go to cloud, and in settings, we have this section called connectors.
00:06:16And we have to click customize.
00:06:19And for this, you will need the tunnel URL.
00:06:21So make sure you have activated the tunnel.
00:06:24And we can just copy this URL.
00:06:26And in the connectors section, you just have to add a custom connector.
00:06:31Give a name.
00:06:33And paste in that tunnel URL.
00:06:38And just click add.
00:06:39I've already added it here.
00:06:41And last thing you need to do is click connect.
00:06:45And you can also set the permissions if you want these tools to run automatically or you want an approval from the user.
00:06:53And then we can just go ahead and start a new chat.
00:06:57And in the new chat, I can now ask, give me some recommendations for the best lenses under $2,000 for portrait shooting.
00:07:05And we can now see that it says it will search the lens shop catalog for some great portrait options under $2,000.
00:07:12And look at that.
00:07:14It opened our MCP app.
00:07:16How cool is that?
00:07:18And we can do the same things over here.
00:07:20We can select some of the lenses over here.
00:07:24And then go to the compare section.
00:07:27And it has this little comparison table.
00:07:32The only thing I noticed is that we have this white space from the previous state of our tool.
00:07:40So this is probably something you can fix in CSS.
00:07:43But everything is looking great.
00:07:46And we also have the option to check out.
00:07:49And we get to the checkout section where we can place our order.
00:07:54So how cool is this?
00:07:57And lastly, and most importantly, to see if your app is ready to be published to the public app stores,
00:08:02we can also run an audit of our app through the developer dashboard and check if we have any issues with it.
00:08:10And basically, that's it.
00:08:11In just a few minutes, we were able to build a production-ready MCP app with SkyBridge.
00:08:16That we could theoretically submit to the AI App Store.
00:08:20So I hope this shows you how easy it is to get up and running with SkyBridge.
00:08:25And honestly, it feels like the future of the modern web.
00:08:28I think we will start seeing more and more of these interactive MCP widget apps inside our chatbots.
00:08:34So it's nice that there are frameworks like SkyBridge out there that are making the process of creating these kinds of apps so much easier for us developers.
00:08:42So there you have it, folks.
00:08:44That is SkyBridge in a nutshell.
00:08:46So what do you think about SkyBridge?
00:08:47Have you tried it?
00:08:48Will you use it?
00:08:49Let us know in the comments section down below.
00:08:52And folks, if you like these types of technical breakdowns, please let me know by smashing that like button underneath the video.
00:08:58And also, don't forget to subscribe to our channel.
00:09:01This has been Andrus from BetterStack, and I will see you in the next videos.
00:09:16you

Key Takeaway

SkyBridge is an open-source TypeScript framework that simplifies MCP app development by providing built-in developer tools, live state synchronization, and instant public tunneling for AI chatbots.

Highlights

  • Model Context Protocol (MCP) apps replace static JSON or markdown outputs with interactive UI widgets rendered directly inside AI chatbots.

  • SkyBridge is an open-source TypeScript framework that handles complex protocol bridging, state synchronization, and security rules for MCP app development.

  • The SkyBridge DevTools control panel includes a local sandbox with live hot module replacement, an integrated public tunneling feature, and a Beacon Audit tool for pre-submission checks.

  • Installing the SkyBridge skill into an AI agent enables automated app generation based on conversational requirements and e-commerce schema specifications.

  • SkyBridge synchronizes application state in real time between user actions in the widget and the underlying large language model context.

Timeline

The Shift Toward Interactive MCP Applications

  • Chatbot users increasingly rely on conversational interfaces for routine searches and research tasks.
  • Applications are moving away from returning raw JSON files or markdown documents in favor of interactive widgets inside chat interfaces.
  • SkyBridge is an open-source framework designed to simplify the creation of custom Model Context Protocol apps.

The growing reliance on AI chatbots creates a paradigm shift in how digital services deliver information. Traditional text responses limit user interaction, prompting a transition toward dynamic UI components. Frameworks like SkyBridge streamline the development process for these interactive tools across various use cases.

Architecture and Development Workflow of MCP Apps

  • MCP apps serve two concurrent users within the same interface: the human user and the AI assistant.
  • User inputs and background data updates synchronize instantly across both the interface and the large language model.
  • SkyBridge manages protocol bridging, state synchronization, and security rules using standard React and TypeScript code.
  • The redesigned DevTools control panel features a local sandbox, an integrated public tunneling option, and a Beacon Audit tool.

Building an MCP app requires supporting both human interaction through clicks and sliders and automated data updates driven by an LLM. SkyBridge abstracts the underlying plumbing required for protocol management and security. Its control panel includes a local sandbox with hot module replacement, a single-click secure tunnel for ChatGPT integration, and an audit utility to catch store rejection triggers.

Building and Testing an E-Commerce MCP App

  • Installing the SkyBridge skill into a repository allows an AI agent to generate functional MCP apps from natural language prompts.
  • The developer dashboard provides local emulation views for desktop and mobile displays, as well as light and dark mode toggles.
  • Selecting products within the widget updates the application state in real time across comparison and checkout views.

Integrating SkyBridge into an existing e-commerce repository enables AI assistants to construct a dedicated MCP app based on specific project requirements. Developers can inspect tool outputs, state parameters, and responsive layouts inside the local emulator dashboard. Actions taken inside the widget, such as selecting camera lenses for comparison, immediately reflect in the application state.

Deploying and Auditing the MCP App

  • Tunnel URLs generated by SkyBridge connect local development servers directly to ChatGPT developer settings as custom connectors.
  • Chatbots can successfully query local catalogs, render interactive comparison tables, and process checkouts via the deployed MCP app.
  • The Beacon Audit tool scans app metadata and security policies to verify readiness for public AI app stores.

Connecting the local server to ChatGPT requires pasting the secure tunnel URL into the platform's connector settings. Once connected, conversational prompts successfully trigger the MCP app to display product recommendations, comparison grids, and checkout forms. Running the built-in audit tool completes the development lifecycle by identifying metadata or security issues prior to public submission.

Community Posts

View all posts