mcpup 0.1.0

Generate Pydantic models for your Python package functions 🐶

🐶 mcpup

Automatically generate Pydantic models for all functions in a Python package.

Features

• Automatic Function Discovery: Scans all modules in a package to find functions
• Pydantic Model Generation: Creates Pydantic models for function parameters using pydantic-function-models
• Validation: Generated models perform validation according to type hints
• Package Structure Preservation: Maintains the original package's module structure
• Optional uv Integration: Can install packages on-the-fly with uv

Installation

# Install with pip
pip install mcpup

# Or with uv
uv pip install mcpup

Requirements

• Python 3.10+
• uv (recommended)

Command Line Usage

Generate Pydantic models for all functions in a package:

mcpup package_name

Options:

--output, -o DIRECTORY       Directory to save generated models [default: ./mcpup_models]
--install, -i                Install the package using uv before generating models
--include-private            Include private functions (starting with underscore)
--module, -m TEXT            Specific modules to include (can be used multiple times)
--help                       Show help message and exit

Examples

Generate models for all functions in the polars package:

mcpup polars

Generate models only for specific modules:

mcpup polars --module dataframe --module series

Install the package first, then generate models:

mcpup some-package --install

Include private functions:

mcpup mypackage --include-private

Programmatic Usage

You can also use mcpup programmatically:

from mcpup.scanner import scan_package
from mcpup.generator import generate_models
from pathlib import Path

# Scan a package for functions
functions = scan_package("mypackage", include_private=False)

# Generate models
output_path = Path("./models")
generate_models(functions, output_path)

Using Generated Models

After generating models, you can use them to validate function arguments:

# Import the generated model
from mcpup_models.mypackage.mymodule import MyFunction

# Validate function arguments
valid_args = MyFunction.model.model_validate({
    "arg1": "value",
    "arg2": 123
})

# Call the function with validated arguments
from mypackage.mymodule import my_function
result = my_function(**valid_args.model_dump(exclude_unset=True))

Contributing

Contributions welcome!

1. Issues & Discussions: Please open a GitHub issue or discussion for bugs, feature requests, or questions.
2. Pull Requests: PRs are welcome!
   • Install the dev extra with pip install -e ".[dev]"
   • Run tests with pytest
   • Include updates to docs or examples if relevant

License

This project is licensed under the MIT License.