modelcontextprotocol/mcpb

Desktop Extensions: One-click local MCP server installation in desktop apps

What it solves

MCP Bundles (MCPB) simplifies the distribution and installation of local Model Context Protocol (MCP) servers. Instead of requiring users to manually configure servers, it provides a standardized zip-archive format (.mcpb) that allows end users to install local AI tools and capabilities with a single click.

How it works

An MCPB file is a zip archive containing a local MCP server and a manifest.json file that describes the server's capabilities and configuration. The project provides three main components:

  1. Bundle Specification: A standard for how the manifest and server files should be structured.
  2. CLI Tool: A command-line interface (mcpb init and mcpb pack) that helps developers create the manifest and package their servers into bundles.
  3. Loading Logic: Code used by applications (like Claude for macOS and Windows) to load, verify, and install these bundles.

Who it’s for

  • MCP Server Developers: Those who want to make their local AI tools easy to distribute and portable across different AI desktop applications.
  • App Developers: Developers of AI applications that want to integrate single-click installation support for local MCP servers.

Highlights

  • Single-Click Installation: Enables a user-friendly installation experience similar to Chrome or VS Code extensions.
  • Cross-Language Support: Supports servers written in Node.js, Python (including UV runtime), and compiled binaries.
  • Cross-Application Compatibility: Designed as an open specification to be used by multiple AI desktop applications beyond just Claude.
  • Developer Tooling: Includes a CLI for rapid bundle creation and manifest generation.

Related

  • Dispatch
    Claude Desktop Extensions release notes / what's new

    Anthropic has introduced Desktop Extensions (.mcpb files), a new packaging format that enables one-click installation of Model Context Protocol (MCP) servers for Claude Desktop, removing the need for manual configuration and developer tools.

  • Project
    wonderwhy-er/DesktopCommanderMCP

    An MCP server that enables AI assistants to search, edit files, and run terminal commands on your local machine, with native support for Office documents and process management.

  • Project
    pathintegral-institute/mcpm.sh

    A CLI tool for managing Model Context Protocol (MCP) servers, providing global configuration, server discovery, and easy integration across multiple AI clients.

  • Project
    microsoft/mcp

    A collection of Model Context Protocol (MCP) servers and core tooling that standardizes how AI agents and LLMs connect to Microsoft services and external data sources.

  • Project
    mondaycom/mcp

    An open framework and MCP server that enables AI agents to securely interact with monday.com to manage projects, track tasks, and execute workflows.