MCP Python Function Generator Server

What is MCP Python Function Generator Server?

An MCP server that exposes a tool to generate Python function code based on a natural language description, with optional human-written test cases for validation.

How to use MCP Python Function Generator Server?

Install the uv package manager, clone or download the repository, run the provided setup script (setup.bat on Windows, setup.sh on Linux/macOS), then configure your MCP client with the command and arguments printed by the script. The server exposes a single tool create_python_function that takes a required requirement string and an optional test_cases array of objects.

Key features of MCP Python Function Generator Server

• Generates Python function code from natural language descriptions.
• Supports optional human-written test cases for validation.
• Parameter names are locked by the first test case's keys.
• Only human‑written tests are allowed; AI-generated tests are not supported.
• Runs locally via the uv Python package manager.
• Automatically validates syntax, function definition, and test results.

Use cases of MCP Python Function Generator Server

• Quickly generate a Python function from a user’s description without tests.
• Generate a Python function that must pass human‑validated test cases.
• Integrate into MCP clients (e.g., Claude Desktop) for on‑demand code generation.
• Use as a teaching or prototyping tool for Python functions.

FAQ from MCP Python Function Generator Server

What dependencies does this server require?

Python and the uv package manager must be installed and available in the system PATH.

Where does the generated code run?

The code runs locally on the user’s machine, within the server’s process.

What transport does the server use?

It uses standard MCP transport (stdio) configured via the client’s settings.

Are test cases generated by an AI allowed?

No. The test_cases argument is intended only for tests written or validated by a human. AI‑generated tests are not supported and may cause incorrect validation.

Does the server support non‑deterministic outputs or floating‑point tolerances?

No. The test comparison uses exact equality (==); non‑deterministic outputs or floating‑point tolerances are not handled automatically.