July 2, 2026

Dev Tools|Index 03

Manufact Offers a Vercel-like Cloud for Interactive AI Chat Apps

Manufact, from the creators of mcp-use, simplifies the deployment, testing, and monitoring of Model Context Protocol (MCP) applications, which enable interactive UIs within LLM clients like ChatGPT and Claude.

Via
AITECH TOKYO Editors
Dateline
Tokyo, July 2, 2026
Date
July 2, 2026
Time
5 min read
Manufact Offers a Vercel-like Cloud for Interactive AI Chat Apps

Tagline

Cloud platform for building and deploying interactive AI chat apps

Who & Why

For a product manager at a SaaS company in Tokyo looking to integrate their service directly into LLM chat interfaces, Manufact streamlines the development and deployment of interactive MCP apps.

vs. Existing

This competes with the manual, complex process of building and deploying MCPs directly to LLM marketplaces, offering a specialized, Vercel-like developer experience.

Tokyo Take

Interesting abroad, but for Tokyo professionals, the practical impact is 12-24 months out, contingent on broader Japanese enterprise adoption of MCP and robust Japanese language model support for interactive UIs.

Manufact is a specialized cloud platform designed for developing, deploying, and monitoring Model Context Protocol (MCP) applications and servers. Positioned as "to mcp-use as Vercel is to Next.js," it aims to streamline the process for development teams integrating their services directly into major LLM clients.

MCP allows companies to embed fully interactive user interfaces and display branding within conversational AI platforms like ChatGPT and Claude. This integration means users can interact with external products and services directly within their chat environment, eliminating the need to switch between applications.

The co-founders, Pietro and Luigi, previously focused on open-source SDKs for building agents using MCP servers. However, with the consolidation of agent frameworks into a few major harnesses, their attention shifted to the server side, recognizing the growing importance of first-class integration with company systems via MCP.

The platform addresses several pain points in the nascent MCP ecosystem: the tricky and manual store submission process, the difficulty in designing effective MCPs beyond simple API proxies, the rapid evolution of the MCP specification, and complex authentication issues. Many companies also remain unaware that MCPs can return interactive UIs.

Manufact simplifies deployment by connecting to a GitHub repository, automatically detecting the framework, and providing a live MCP URL. This URL is used for immediate testing and debugging within a dedicated chat environment. Preview deployments support testing experimental branches.

For quality assurance, Manufact offers automated tests that install and verify the MCP server within actual ChatGPT and Claude clients, ensuring functionality where users engage. It also provides a pre-submission check against store requirements, helping developers avoid rejection.

Post-deployment, the platform offers MCP-specific analytics, detailing user engagement, tool call volumes, and client sources. This granular data helps teams understand how their integrated services are being utilized within AI chat environments. Manufact offers usage-based pricing with free credits for new users.

If AI apps (Codex, Claude Desktop) are the new browsers, as PG said in a recent tweet, then MCPs are the new websites.

This shift implies that products will be organically discovered and interacted with directly within chat interfaces, fundamentally changing how users engage with digital services and how companies distribute their offerings.

The Briefing

World AI tech, read from Tokyo. Once a week, in Japanese.

Each Friday: the five global AI tech stories Japanese business professionals should know about this week, translated and read through a Tokyo lens — what it means for Japan, what to act on, what to keep watching.

We respect your inbox. Unsubscribe anytime.