Skip to main content
MetaMCP DiagramMetaMCP is a MCP proxy that lets you dynamically aggregate MCP servers into a unified MCP server, and apply middlewares. MetaMCP itself is a MCP server so it can be easily plugged into ANY MCP clients.

🎯 Use Cases

Group & Host MCP Servers

Group MCP servers into namespaces, host them as meta-MCPs, and assign public endpoints (SSE or Streamable HTTP) with authentication.

Cherry-pick Tools

Pick only the tools you need when remixing MCP servers. Apply pluggable middleware for observability and security.

Enhanced Inspector

Use as enhanced MCP inspector with saved server configs, and inspect your MetaMCP endpoints.

Context Engineering Engine for MCP

Use as Elasticsearch for MCP tool selection (coming soon).
Generally developers can use MetaMCP as infrastructure to host dynamically composed MCP servers through a unified endpoint, and build agents on top of it.
MetaMCP Screenshot

Quick Start

Get MetaMCP running in minutes with Docker Compose:
git clone https://github.com/metatool-ai/metamcp.git
cd metamcp
cp example.env .env

Continue with Quickstart Guide

Follow our complete setup guide to configure your first MCP servers and namespaces.

Core Concepts

Understanding these key concepts will help you get the most out of MetaMCP:

Integrations

Connect MetaMCP to your favorite tools and platforms:

Features

  • Better Auth for frontend & backend
  • Session cookies for secure connections
  • API key authentication for external access
  • OpenID Connect (OIDC) support for enterprise SSO
  • Multi-tenancy with public/private scopes
  • SSE (Server-Sent Events) for MCP backward compatibility and real-time connections
  • Streamable HTTP for Streamable HTTP-based MCP communication (now standard remote MCP connection)
  • OpenAPI endpoints compatible with clients like Open WebUI
  • STDIO compatibility via proxy for local
  • Built-in support for English and Chinese
  • Easy to add additional languages
  • Locale-based routing and content
  • Pre-allocated idle sessions for reduced cold start time
  • Configurable session management
  • Designed for at least 2GB-4GB memory instances
  • Docker-based deployment with nginx support
  • Cluster scaling and separation of MCP manager and worker will release in the future.

Getting Help