pymcp

pymcp is a Python-based MCP server template designed to work with FastMCP-style tooling. It exposes a suite of built-in tools that can be invoked by clients to perform common tasks, such as greeting, generating secure passwords, performing web searches, combinatorial calculations, executing Python code securely, text summarisation in pirate style, and generating random values from specialized distributions. Clients can call tools like greet, generate_password, text_web_search, permutations, run_python_code, pirate_summary, and vonmises_random, each with well-defined inputs and structured outputs. The server is intended to be run using uv (Python/uv) and can be installed from a Python package. The included environment variables section highlights optional runtime configuration options that influence how the server communicates with external services or handles input/output. Use the tools to build interactive MCP-powered experiences for Minecraft-related tooling, automation, or research tasks.

Prerequisites:

• Python 3.12 or newer
• uv (Python package runner) installed
• Just (build tool) installed

Install and run locally (minimal dependencies):

1. Ensure Python is installed and available in your PATH.
2. Install uv and just if not already installed, following their official guidance.
3. Clone or download this repository into your working directory.
4. In the working directory, install the minimal dependencies and set up the environment:

just install

5. Run the MCP server using uvx (Python/uv) pointing to the pymcp package or entry point as configured in your environment:

uvx pymcp

Optional: If you prefer a full development environment with all dependencies, switch the tag in the Justfile to install-all and run the same commands.

Notes:

• Ensure you meet Python version requirements (3.12+ as indicated).
• If you clone this repo as a template, you may need to adjust the uvx target to the actual package/module name you provide in your project.

Environment variables and configuration options may vary by deployment. Typical considerations include:

• PYMC (environment variable placeholder): Used for runtime configuration in some setups; replace with actual variable names your deployment expects.
• If your deployment uses external services (e.g., web search proxy), you may need to set DDGS_PROXY or other proxy-related environment variables as indicated by the tool descriptions.
• Ensure the Python environment has access to the required dependencies and that the MCP tools are exposed to clients with appropriate authentication/authorization if needed.
• When running locally, you can test each tool individually to verify inputs/outputs before integrating into larger workflows.