yamcp
Organize your MCP servers in local workspaces, share them as Yet-Another-MCP through a single command
claude mcp add --transport stdio hamidra-yamcp npx -y yamcp
How to use
YAMCP (Yet Another MCP) acts as a local workspace manager that aggregates multiple MCP servers into a single, unified gateway. It helps you organize servers into workspaces (YAMs) tailored to specific AI apps or workflows, and exposes them through a gateway that AI clients can connect to via MCP. With YAMCP, you can import server configurations, create and manage workspaces, and run a gateway that centralizes logging and monitoring for all included MCP servers. This makes it easier to scale your MCP ecosystem and provide a consistent interface to the AI applications you use. The tool supports both local servers (via stdio) and remote servers (via SSE), and it provides commands to manage providers, workspaces, and the gateway runtime from a single CLI.
Once set up, you can import a set of MCP servers into a workspace, spin up the gateway for that workspace, and point your AI app’s MCP client at the gateway. The gateway handles routing requests to the appropriate MCP servers within the workspace and collects logs in a central store for debugging and auditing. You can create separate workspaces for different tasks (e.g., coding, design, data science) and link them to different AI apps or workflows, enabling clean separation and straightforward management.
How to install
Prerequisites:
- Node.js (LTS version recommended) and npm installed on your system
- Basic familiarity with using a command line
-
Install Node.js and npm (if not already installed)
macOS/Linux: curl -fsSL https://fnm.vercel.app/install.sh | bash exec $SHELL nvm install --lts
Windows (PowerShell): choco install nodejs-lts
-
Install YAMCP globally (or you can use npx directly as shown in the config)
npm install -g yamcp
or run directly via npx without global install:
npx yamcp
-
Verify installation
yamcp --version
-
Optional: Prepare a sample config and import servers
- Create a JSON file with your MCP server configurations (see example below)
- Use the CLI to import or add servers as needed
Example usage to start with a sample workspace:
yamcp server add yamcp yam create yamcp run <workspace-name>
Additional notes
Tips and common issues:
- Ensure your Node.js version is compatible with yamcp. If you encounter permission errors with npm global installs, consider using a Node version manager (e.g., nvm) and avoid sudo.
- When importing servers, make sure the MCP server endpoints are reachable and that any required authentication or network access is configured.
- The gateway centralizes logs; if you don’t see logs, check the store location configured by yamcp and ensure write permissions.
- Workspaces (YAMs) can be named intuitively (e.g., coding-yam, design-yam). Use descriptive names to avoid confusion when managing multiple workspaces.
- If running via npx, network access is required to fetch the yamcp package; ensure your environment allows network access to npm registries.
- For remote MCP servers, verify SSE connectivity and any security requirements (TLS, tokens) before adding them to a workspace.
Related MCP Servers
mcp-router
A Unified MCP Server Management App (MCP Manager).
agent
A unified Model Context Protocol server implementation that aggregates multiple MCP servers into one.
pluggedin-app
The Crossroads for AI Data Exchanges. A unified, self-hostable web interface for discovering, configuring, and managing Model Context Protocol (MCP) servers—bringing together AI tools, workspaces, prompts, and logs from multiple MCP sources (Claude, Cursor, etc.) under one roof.
mcp-bundler
Is the MCP configuration too complicated? You can easily share your own simplified setup!
mcp -zotero-dev
Give your AI assistant superpowers for Zotero plugin development. 25 tools for screenshots, DOM inspection, JavaScript execution, build integration, and debugging via Model Context Protocol.
mcp-more
A modern desktop application for managing Model Context Protocol (MCP) servers.