---
title: "Deploy MCPHub"
description: "MCPHub 1.0: one endpoint and dashboard for all your MCP servers."
category: "AI/ML"
url: https://railway.com/deploy/mcphub
---

# Deploy MCPHub

MCPHub 1.0: one endpoint and dashboard for all your MCP servers.

**[Deploy MCPHub on Railway](https://railway.com/template/mcphub)**

Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/mcphub/manifest.json

- **Creator:** Agaz Self-Host
- **Category:** AI/ML

## Template content

### mcphub https://raw.githubusercontent.com/samanhappy/mcphub/v1.0.40/docs/favicon.svg

- **Image:** samanhappy/mcphub:1.0.40
- **Health check:** /health
- **Public domain:** Yes

## Documentation

# Deploy and Host MCPHub on Railway

MCPHub is a hub for Model Context Protocol servers. It runs and manages many MCP servers, stdio or remote, from one dashboard and exposes them through a single Streamable HTTP or SSE endpoint, with groups, per-server routes, smart tool routing, bearer keys and users. AI clients connect once instead of configuring every server.

## About Hosting MCPHub

This template runs the official `samanhappy/mcphub:1.0.40` image as one service. The bootstrap `admin` user gets a generated `ADMIN_PASSWORD` (the upstream default `admin123` does not work), the dashboard API needs a login, and the MCP endpoints need a bearer key you create in the dashboard. Server settings, users and keys are stored in `mcp_settings.json` on a Railway volume at `/app/data`, so they survive redeploys. Stdio servers are started with `npx` or `uvx` inside the container, so they share its CPU and memory. It fits the Hobby plan for a handful of servers.

## Common Use Cases

- One MCP endpoint for Claude, Cursor and other AI clients
- Hosting stdio MCP servers remotely for a team
- Grouping tools per project with separate endpoints and keys

## Dependencies for MCPHub Hosting

- `samanhappy/mcphub:1.0.40` (official image)
- A Railway volume at `/app/data`

### Deployment Dependencies

- [MCPHub on GitHub](https://github.com/samanhappy/mcphub)
- [MCPHub v1.0.40 release](https://github.com/samanhappy/mcphub/releases/tag/v1.0.40)
- [Model Context Protocol](https://modelcontextprotocol.io/)

### Implementation Details

| Service | Source | Networking | Storage |
| --- | --- | --- | --- |
| mcphub | `samanhappy/mcphub:1.0.40` | public domain on 3000 | volume at `/app/data` |

Connect a client to all enabled servers:

```json
{ "mcpServers": { "hub": { "url": "https:///mcp", "headers": { "Authorization": "Bearer " } } } }
```

| Variable | Default | Purpose |
| --- | --- | --- |
| `ADMIN_PASSWORD` | generated | Password for the bootstrap `admin` user |
| `JWT_SECRET` | generated | Signs dashboard sessions |
| `INSTALL_BASE_URL` | `https://` | Public URL shown in install snippets |

Notes:

- Create bearer keys in Settings → Keys; keys with access type "all" also work for the dashboard API.
- Use `/mcp/` or `/mcp/` for narrower endpoints.
- The container runs as root upstream; only add MCP servers you trust.

This is a community-maintained deployment package and does not imply affiliation with or endorsement by the MCPHub project.

## Why Deploy MCPHub on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying MCPHub on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.


## Similar templates

- [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform.
- [stella](https://railway.com/deploy/stella) — Self-host stella with web, API, Postgres, Redis, and object storage.
- [Hermes Agent | OpenClaw Alternative with Dashboard](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-wit) — Self-Hosted Hermes AI Agent for Telegram, Discord & Slack

Open this page in a browser: https://railway.com/deploy/mcphub
