- Home
- MCP servers
- Multiverse
Multiverse
- typescript
77
GitHub Stars
typescript
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"lamemind-mcp-server-multiverse": {
"command": "npx",
"args": [
"-y",
"@lamemind/mcp-server-multiverse@latest",
"/path/to/your/job-multiverse.json"
]
}
}
}The Multiverse MCP Server lets you run several isolated MCP server instances side by side, each in its own universe with separate configurations, files, and function namespaces. This enables parallel development and testing of identical server types without cross-talk between contexts or projects.
How to use
You run multiple MCP server universes from a single control point. Each universe starts its own server instance and exposes a distinct set of functions, so you can work on several projects that use the same MCP server type without interference. You can configure one universe for a job or production workflow and another for a side project, each with its own root path, environment, and resources.
How to install
Prerequisites you need before installation are Node.js and npm, plus the Claude Desktop app if you plan to manage multiverses from a GUI. Ensure Node and npm are accessible from your command line.
Install Node.js and npm if you don’t have them yet. Use your system package manager or download from the official site and follow the installation prompts.
Install Claude Desktop and prepare the configuration file to define how many multiverse servers you want to run. Create or edit the configuration file path where Claude Desktop can discover your multiverse definitions.
Add the multiverse server entries to your configuration so Claude Desktop can start the universes automatically when the app starts. Use the following configuration as a template.
{
"mcpServers": {
"job_multiverse": {
"command": "npx",
"args": [
"-y",
"@lamemind/mcp-server-multiverse@latest",
"/path/to/your/job-multiverse.json"
]
},
"side_proj_mv": {
"command": "npx",
"args": [
"-y",
"@lamemind/mcp-server-multiverse@latest",
"/path/to/your/side-project-multiverse.json"
]
}
}
}
Configuration overview
The multiverse configuration defines separate server worlds, each with its own command to launch, optional environment variables, and a path to the universe JSON that describes the server instances inside that universe.
Examples and scenarios
Create two isolated instances of the same MCP server type to keep distinct databases or namespaces separate. You can label each universe clearly and point it to different resource paths or configurations.
Automatic server restart
Enable file watching during development so a change in a universe’s files triggers a graceful restart. This helps you iterate quickly without manual restarts.
Hiding and disabling servers
You can hide or disable specific servers inside a multiverse. Hiding prevents those functions from appearing in the MCP client, while disabling a server stops it from starting at all until you re-enable it.
Full multi-universe example
A single multiverse setup can include multiple server types, each with its own runtime and resources. This enables complex development environments where different projects share the same MCP server framework but operate in complete isolation.
Notes
Only commands and URLs shown in explicit code blocks or clearly labeled examples should be used as MCP endpoints. Avoid guessing endpoints from badges or previews. Configure your universes with precise, explicit paths and commands.
Available tools
Run Multiple Instances
Start and manage several MCP server universes from a single control point, each operating in its own isolated namespace.
Automatic Server Restart
Automatically restart a universe when changes are detected in its files, reducing manual steps during development.
JSON-based Configuration
Define every universe with a JSON configuration that specifies command, args, environment, and file watching settings.