mcpField
M
mcp-minimal

mcp-minimal

Local
github.com/janspoererDeveloper Tools0Updated 1w ago
Repository

About mcp-minimal

mcp-minimal is a minimal MCP server for checking whether MCP client configurations work. Its README provides a pip installation command and links to setup instructions for Claude Desktop on Windows with WSL.

Key features

  1. 01Minimal MCP server
  2. 02Configuration validation

Use cases

  1. 01Check whether an MCP client configuration works
  2. 02Test a Claude Desktop setup with WSL

MCP support

Provides a minimal MCP server that can be used to validate MCP client configurations.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

mcp-minimal

A minimal MCP server to validate that MCP client configurations are working.

Version pinning: action needed

Please address this soon. This repository does not fully pin its versions.

What is missing here:

  • No exact Python version pinned (only the range requires-python = ">=3.10"). There is no .python-version.
  • Unpinned dependencies in requirements.txt (1): fastmcp.

Why this is a problem: without exact pins, every fresh install resolves the newest versions available that day. The same commit can then behave differently on a laptop, in the Docker build and on the server, and a rebuild can break without any code change. Example: on 2026-10-21 a fresh install pulled mcp 2.x and broke several MCP servers because mcp was not pinned anywhere. Developer machines also differ from the servers (a developer PC runs Omarchy/Arch with Python 3.14 and Node 26; dev-playground, hd24devsrv, gptgateway and montage run Ubuntu 24.04 with Python 3.12; mdm, cma-dev and 10.24.241.50 run Ubuntu 26.04 with Python 3.14).

To do:

  • Pin Python: uv python pin <the target server's version>, and commit .python-version.
  • Lock dependencies: uv lock and commit uv.lock, or pin every requirement with == (e.g. uv pip compile requirements.in -o requirements.txt).

Good practice for every HD24 repository:

  • Commit lockfiles (uv.lock, package-lock.json) and treat pin changes like code changes: no silent upgrades.
  • Keep the pinned versions identical to the Docker base image (FROM python:3.12-slim, FROM node:22-slim) and the target server.
  • Tag every Docker base image with a version, never latest.
  • Test locally the way the server runs it: same Dockerfile/Compose and a local .env, not a native python app.py.
  • Test host scripts (apt, systemd, cron) in an ubuntu:24.04 / ubuntu:26.04 container that matches the target server.
  • A local test is a prerequisite, not proof: the final check happens on the target server via the approved deploy path (pull_build_start_container.sh).

Remove this section once everything above is done.

pip install -r requirements.txt

Detailed setup instructions for using Claude Desktop on Windows 10 with WSL (2): https://janspoerer.github.io/posts/.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.