mcpdog
đ Universal MCP Server Manager - Configure once, manage multiple MCP servers through a single interface. Perfect for Claude Desktop, Claude Code, Cursor, Gemini CLI & AI assistants. Web dashboard, auto-detection, unified proxy layer.
claude mcp add --transport stdio kinhunt-mcpdog npx mcpdog@latest
How to use
MCPDog is a universal MCP server manager that lets you run and orchestrate multiple MCP servers through a single interface. It acts as a proxy layer, allowing clients like Claude Desktop, Cursor, or other MCP-enabled tools to interact with several underlying MCP servers as if they were one consolidated endpoint. With MCPDog, you gain a web dashboard for managing servers, smart routing of tool calls, auto-detection of transport protocols, and centralized monitoring so you can see which tools are available and their status in real time. The system supports multiple transports (stdio and HTTP-based streams) so you can connect locally or remotely depending on your deployment.
Once MCPDog is running, you configure your client to point at MCPDog and then add the individual MCP servers you want to use. The dashboard helps you enable or disable specific tools, adjust timeouts, and generate client configurations that you can copy into your applications. This makes it easy to consolidate diverse MCP backendsâfrom lightweight local tools to remote HTTP servicesâbehind a single, consistent interface.
How to install
Prerequisites:
- Node.js v18 or newer
- npm or yarn
- Basic knowledge of MCP servers you plan to run through MCPDog
Step 1: Install MCPDog (via npx as shown in the README or install globally if preferred)
- Ensure you have Node.js installed (v18+)
- From a project directory or your environment, run:
# Install and start MCPDog via npx as an initial setup
npx mcpdog@latest
Step 2: Start MCPDog locally
- If using the default local setup, simply run the command above and the server will start, exposing a dashboard at http://localhost:38881 and an API/interface for managing MCP servers.
- Ensure your environment permits network access to the dashboard and any configured MCP servers.
Step 3: Optional persistent configuration
- Create a configuration file (e.g., ~/.mcpdog/mcpdog.config.json) to persist server definitions and dashboard settings.
- Example configuration (simplified):
{
"mcpServers": {
"mcpdog": {
"command": "npx",
"args": ["mcpdog@latest"]
}
}
}
Step 4: Access the web dashboard
- Open http://localhost:38881 in your browser and start adding MCP servers, configure transports, and monitor status from the dashboard.
Additional notes
Tips and common issues:
- Ensure Node.js is version 18 or newer to meet MCPDog's requirements.
- If you run MCPDog in Docker or cloud environments, map the config directory to a persistent path (e.g., ~/.mcpdog) so your server configurations survive restarts.
- For local development, MCPDog auto-detects and configures transports; if a server isnât reachable, check the transport settings and network accessibility.
- When using HTTP/streamable transports, verify that port mappings (e.g., 38881 for dashboard and 4000 for HTTP streams) are open and not blocked by a firewall.
- Use the dashboard to enable/disable individual servers, monitor tool availability, and generate client configurations for your MCP clients.
Related MCP Servers
taskflow
A task management Model Context Protocol (MCP) server that helps AI assistants break down user requests into manageable tasks with subtasks, dependencies, and notes. Enforces a structured workflow with user approval steps.
cursor-feedback-extension
Save your Cursor monthly quota! Unlimited AI interactions in one conversation via MCP feedback loop.
mongo
MCP server that provide tools to LLMs such as claude in cursor to interact with MongoDB
attio
Model Context Protocol server for Attio CRM - enables Claude Desktop and Cursor to interact with your Attio workspace
mermaid-live
MCP server from iishyfishyy/mermaid-live-mcp
voicepeak
MCP server for VOICEPEAK text-to-speech synthesis