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., "@Basic MCP Serverecho Hello, world!"
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.
Minimal MCP Server Example (fastmcp)
This repo is a minimal, working MCP server example exposing two tools via fastmcp:
echo(text): returns text unchangednow(utc=True): returns current time in ISO 8601
It’s designed as a small, copy‑pasteable starting point.
Contents
server.py: minimal MCP server with two toolsDockerfile: container image for running the serverrequirements.txt: Python dependency list
Related MCP server: Simple MCP Server
Prerequisites
Python 3.11+ (3.12 recommended) or Docker
Run Locally (Python)
Run with Docker
Build the image (tag matches the example config below):
Run it (stdio transport):
Codex CLI / MCP Client Config
Configure your Codex MCP client (e.g., ~/.codex/config.toml) with either Docker or direct Python.
Docker-based (uses the image built above):
Direct Python (runs from your checkout):
How It Works
The server is defined in
server.pyand usesfastmcp.FastMCP("basic")with stdio transport.Tools are registered via
@mcp.tooldecorators and exposed to the client.
Extending
Add new tools by defining functions with type hints and decorating with
@mcp.tool.Keep the interface simple and return JSON-serializable values.
Troubleshooting
If the client can’t start the server, verify
fastmcpis installed (pip show fastmcp).For Docker, ensure the image is built (
docker images | grep mcp-basic) and thatdocker run --rm -i mcp-basic:latestworks.
License
None specified; use as an example and adapt as needed.