
No clickbait detected — the title and thumbnail deliver what they promise.
AI Opinion
The episode convincingly argues that MCP Apps offer a significant advancement in application distribution, particularly by moving beyond text-based interactions and enabling richer user experiences within chat platforms—a point supported by demonstrable adoption across several major services. However, the comparison of MCP Apps’ market potential to the early Apple App Store, while striking, rests on an implied equivalence between nascent app store users and ChatGPT's massive but differently engaged audience; a more nuanced analysis would be helpful. Listeners should also independently verify the claim that OpenAI actively recommends MCP apps for development, as official documentation may offer a different perspective.
Avatars are AI rewrites of the same facts — style changes, not substance.
Summary
The episode explores MCP Apps, a protocol designed to address limitations in current chat-based communication and unlock new possibilities for application distribution. Speakers Ido Salomon and Liad Yosef highlight that relying on text-based interactions diminishes brand identity and user experience, proposing instead an approach where applications deliver interactive user interfaces directly within chat environments via the open protocol MCPUI. This has led to rapid adoption across platforms like ChatGPT, VS Code, Shopify, and GitHub, fostering a growing community building plugins and integrations. A key feature is standardization through "view tools," enabling hosts to interact with applications on behalf of users, while MCP Apps remain agnostic to how UIs are generated. With an estimated 800 million weekly ChatGPT users – representing a market significantly larger than the early Apple App Store – MCP Apps offer developers a “write once, run everywhere” capability for universal compatibility and substantial growth potential.
Avatars are AI rewrites of the same facts — style changes, not substance.
Key Points
Standardization of App Interaction with View Tools
MCP Apps standardizes the interaction between applications and hosts, extending beyond traditional user-initiated actions. This is achieved through 'view tools,' which allow the host (like a chat interface) to directly interact with an application on behalf of the user – for example, filling out forms automatically. The specification for view tools is currently in development and expected to be released shortly.
Generative UI Spectrum and MCP Apps Agnosticism
MCP apps operate within a spectrum of user interface generation, ranging from predefined UIs (like black box iframes) to declarative UIs (JSON render or A2UI) where the chat builds the UI based on instructions, and finally to fully generative UIs. MCP apps are designed to be agnostic to how the UI is generated, meaning they can support various methods including those used in cloud applications where the host can request a UI generation from the app itself.
Massive Market Potential of MCP Apps
The adoption rate of ChatGPT, with its 800 million weekly users (approximately 10% of the global population), demonstrates a significantly accelerated growth compared to the early days of the web. This rapid expansion represents an addressable market roughly 170 times larger than that of the Apple App Store at launch, highlighting the immense potential for MCP apps as a new distribution method.
Write Once, Run Everywhere: Universal Compatibility
A key benefit of MCP Apps is their universal compatibility; an application built once can function across various platforms like Libra Chat and ChatGPT. This 'write once, run everywhere' capability simplifies development and distribution, ensuring broad reach and consistent user experience regardless of the underlying platform.
The Limitation of Text-Based Communication in MCP
Ido and Adi explain that the reliance on text-based communication within MCP is a significant blocker for companies wanting to build MCP servers. They argue that reducing services to textual databases diminishes brand identity and wastes effort invested in user experience design, as it results in unappealing interfaces. The core issue lies in the inefficiency of conveying information through walls of text.
MCP Apps Enable Interactive User Interfaces
The speakers propose a solution where applications send their user interfaces directly to chat environments. This allows users to interact with services like Hugging Face within the chat, rather than viewing static textual responses. This approach aims to preserve brand identity and provide a more engaging experience for users by integrating interactive UI chunks into the MCP environment.
MCPUI as an Open Protocol for Interactive Applications
Ido introduces MCPUI, described as an open protocol for interactive applications over MCP. It not only defines how UI is transmitted but also establishes a standardized method for that UI and the application to communicate with the host system. This allows for interaction between users and services within the chat environment while maintaining control of the flow by the host.
Rapid Adoption and Community Growth Around MCP Apps
MCP apps have seen rapid adoption across various platforms, including ChatGPT, VS Code, Shopify, 11 Labs, Postman, Goose (now integrated into Block’s agentic commerce solution), Cursor, and GitHub. This widespread adoption is further fueled by a growing community building plugins, integrations, and educational courses around the MCP apps protocol, demonstrating its increasing importance in the ecosystem.
Chapters
Claims & Fact Check
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.
Companies don't want to be reduced to a textual database.
ChatGPT apps are based on MCP apps and OpenAI actually recommend using MCP apps as the protocol to build ChatGPT apps.
Block released their agentic commerce solution that is based on MCP apps.
ChatGPT has 800 million weekly users.
The web took around 13 years to reach 800 million users.
MCP apps have a market that is 170 times the size of the Apple App Store at launch.
Was this digest good?
More from AI Engineer

MCP Tasks (async): Why Aren't Any Agents Supporting Them? — Cornelia Davis, Temporal
Aug 2, 2026

When Will The Benchmaxxing Plague End? — Nick Heiner, Surge AI
Aug 2, 2026

Teaching AI to Find Real Vulnerabilities — David Brumley, Bugcrowd
Aug 1, 2026

Rethinking Environments for Long-Horizon Work — Rayan Garg, Theta Software
Aug 1, 2026
Digest any single YouTube video — free.
3 free digests — no card, no sign-up wall.
Or just swap the domain of any YouTube link → instant digest