Your First MCP Server
This tutorial will guide you through adding your first MCP server to Plugged.in. MCP servers extend your AI's capabilities by providing access to tools, resources, and integrations.
- Plugged.in account (free tier is sufficient)
- Basic understanding of the command line (for custom servers)
- Node.js installed (optional, for custom servers)
1Understanding MCP Servers
Before we add our first server, let's understand what MCP servers are and why they're powerful.
MCP (Model Context Protocol) servers are programs that provide additional capabilities to AI assistants. Think of them as plugins that extend what your AI can do.
- Tools: Functions the AI can call (e.g., read files, search web, send emails)
- Resources: Data the AI can access (e.g., documents, databases, APIs)
- Prompts: Pre-configured instructions for specific tasks
2Navigating to MCP Servers
Let's start by navigating to the MCP Servers section in Plugged.in.
Screenshot: Plugged.in dashboard with MCP Servers menu highlighted
The MCP Servers option in the sidebar navigation
3Adding Your First Server
Now let's add your first MCP server. You can choose between official servers or custom ones.
- Click the 'Add Server' button
- Browse through the official server catalog
- Select 'Filesystem' server (a good starter server)
- Click 'Install' to add it to your profile
Screenshot: Official server catalog with Filesystem server
The official server catalog showing various MCP servers
4Configuring Your Server
Each server can be configured with specific settings and environment variables.
Configure how your server runs:
{
"name": "Filesystem Server",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/allowed/directory"],
"env": {
"NODE_ENV": "production"
}
}
Some servers require API keys or other configuration through environment variables.
5Testing Your Server
Let's make sure your server is working correctly.
- Save your server configuration
- Click the 'Test Connection' button
- Wait for the connection to establish
- Check the discovered tools and resources
Screenshot: Successful server connection with tools listed
A successfully connected server showing available tools
Common Issues & Solutions
If your server fails to connect:
- Check that all required fields are filled
- Verify environment variables are set correctly
- Ensure the server command/path is correct
This usually means the server connected but isn't configured properly. Check the server's documentation for required setup.