mcp-use/mcp-use

Project overview

A fullstack Model Context Protocol (MCP) framework providing SDKs, a web inspector, and scaffolding tools for building MCP servers and apps. It supports self-hosted local execution and requires external LLMs for agent functionality.

Project type

MCP · AI Agent · Infrastructure

Use cases

Automation

Deployment

Refer to project documentation

License

MIT

Best for

• Developers and AI engineers who need a self-hosted, locally runnable framework with built-in debugging and scaffolding tools to build and test MCP servers and interactive apps.

Key capabilities

• Provides a full MCP Agent and Client implementation to connect LLMs to MCP servers.
• A web-based previewer and debugger to test and debug MCP servers interactively (auto-included or standalone).
• A build tool with hot reload and auto-inspector functionality.
• A tool to quickly scaffold a new MCP server or app project.

Limitations and risks

• Managed hosting and observability features depend on the external Manufact MCP Cloud platform.

Getting started

• Setup difficulty is documented as easy. First success paths include installing via npm/pip or using npx create-mcp-use-app to scaffold a project via command line interface.

Evidence and sources

• GitHub project description: The fullstack MCP framework to develop MCP Apps for ChatGPT / Claude & MCP Servers for AI Agents.
• README: Deploy on Manufact MCP Cloud: Connect your GitHub repo and have your MCP Server and App up and running in production with observability, metrics, logs, branch-deployments, and more
• README: | Package | Description | Version | | ---------------------- | ----------------------------------------------- | ------------------------------------------------------------------…
• README: MCP Apps let you build interactive widgets that work across Claude, ChatGPT, and other MCP clients — write once, run everywhere.
• README: Build your first MCP Server or MPC App: ```bash npx create-mcp-use-app@latest ```

Reviewed at: 2026-07-18T03:08:34Z