About
Mcphub centralizes, monitors, and orchestrates multiple Model Context Protocol servers through a single Streamable HTTP endpoint, offering real‑time dashboards, group access control, and dynamic configuration without downtime.
Capabilities

Overview
MCPHub is a unified gateway for managing multiple Model Context Protocol (MCP) servers in one place. It solves the common pain point of juggling several independent MCP endpoints by providing a single, streamable HTTP interface that routes requests to any configured server. Developers can now expose diverse tools—such as web browsers, API wrappers, or custom workflows—to AI assistants without having to write bespoke adapters for each one. The result is a scalable, plug‑and‑play ecosystem where new MCP services can be added or removed on the fly without downtime.
The server’s core value lies in its flexibility and centralization. MCPHub supports both the standard stdio protocol used by many local servers and the SSE (Server‑Sent Events) protocol that allows real‑time streaming of responses. Through a sleek web dashboard, operators can monitor server health, view performance metrics, and adjust configuration in real time. Grouping servers into logical collections enables fine‑grained permission control: only users with the right role can access specific tool sets, making it ideal for enterprise deployments where data sensitivity matters.
Key capabilities include:
- Hot‑Swappable Configuration: Add, remove, or update MCP servers via a JSON file or the dashboard—no service restarts required.
- Smart Routing (experimental): A vector‑based semantic search engine indexes all available tools, automatically suggesting the most relevant ones for a given user prompt. This reduces manual tool selection and speeds up workflow integration.
- Role‑Based Access Control: Built‑in JWT authentication with bcrypt password hashing lets you create multiple user accounts and assign permissions at the group level.
- Docker‑Ready Deployment: The entire stack is containerized, simplifying production rollout and scaling.
Real‑world scenarios where MCPHub shines include: a research lab that needs to expose several web‑scraping and data‑fetching MCP servers to an AI assistant; a SaaS platform that wants to offer customers a single endpoint for all its custom tools; or an internal tooling hub where developers can quickly spin up new MCP services and grant selective access to their teams. By consolidating these servers into one coherent interface, MCPHub eliminates configuration drift, reduces operational overhead, and accelerates the time to value for AI‑powered applications.
Related Servers
MarkItDown MCP Server
Convert documents to Markdown for LLMs quickly and accurately
Context7 MCP
Real‑time, version‑specific code docs for LLMs
Playwright MCP
Browser automation via structured accessibility trees
BlenderMCP
Claude AI meets Blender for instant 3D creation
Pydantic AI
Build GenAI agents with Pydantic validation and observability
Chrome DevTools MCP
AI-powered Chrome automation and debugging
Weekly Views
Server Health
Information
Explore More Servers
Cursor Chat History Vectorizer & Dockerized Search MCP
Turn Cursor chat history into searchable embeddings
Pearl MCP Server
Bridge AI and human experts through a unified protocol
GitHub MCP Server
Secure AI‑driven IDE integration via Docker
Storyblok MCP Server
Unified CMS management via Model Context Protocol
Yourware MCP
Upload projects to Yourware with a single command
MCP-MongoDB-MySQL-Server
Unified MySQL and MongoDB MCP server for AI models