MCP Server
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP Serverwhat's the weather in London right now?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server (FastMCP)
A minimal Model Context Protocol server using FastMCP, with one tool, one prompt, and one resource.
Type | Name | Description |
Tool |
| Safe math expression evaluation ( |
Prompt |
| Ask the model to explain a topic in simple terms |
Resource |
| Server name and capabilities (read-only) |
Requirements
Python 3.10+
pip
Setup
1. Create and activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate2. Install dependencies
pip install -r requirements.txt3. Run the server
MCP servers use stdio by default. Your MCP client launches the process. To verify it starts:
python server.py
# Blocks waiting for MCP messages. Press Ctrl+C to quit.Connecting to Claude Desktop
Add to your Claude Desktop config:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"example-server": {
"command": "/absolute/path/to/MCP-Server/.venv/bin/python",
"args": ["/absolute/path/to/MCP-Server/server.py"]
}
}
}Use your actual paths and restart Claude Desktop.
Using with MCP Dashboard
The MCP Dashboard launches this server as a subprocess and talks to it over stdio, so tools, resources, and prompts all appear in the UI.
From the MCP-Dashboard directory (with Flask installed), run:
# Unbuffered (-u) so the dashboard receives list responses immediately
python app.py -- python -u path/to/MCP-Server/server.pyWindows:
python app.py -- python -u path\to\MCP-Server\server.pyUse a Python that has both flask and fastmcp installed (or use the full path to this repo's venv Python if you use one).
The dashboard will open at http://localhost:8080 and show the calculate tool, app://info resource, and explain prompt. You can try the tool, read the resource, and get the prompt from the browser.
Reference
Tool: calculate
Input:
{ "expression": "sqrt(2) * pi" }Output:
"4.442882938158366"Supports:
+,-,*,/,**,%,//, and functions likeabs,round,sqrt,ceil,floor,log,log10,sin,cos,tan; constantspi,e.
Prompt: explain
Input:
{ "topic": "recursion" }Returns a user message asking the model to explain the topic in simple terms.
Resource: app://info
Read-only. Returns JSON:
{"name": "mcp-server", "capabilities": ["calculate", "prompts", "resources"]}.
Project structure
MCP-Server/
├── server.py
├── requirements.txt
└── README.mdTo add more tools, prompts, or resources, use the @mcp.tool, @mcp.prompt, and @mcp.resource(...) decorators in server.py.
This server cannot be installed
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/michaelmoser8875/MCP-Server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server