Release v2.7.5. (Tutorial)

License PyPI Version Python Versions Documentation Tests Downloads

The cmdkit library implements a few common patterns needed by well-formed command-line applications in Python. It only touches a few concepts but it implements them well. The idea is to reduce the boilerplate needed to get a full featured CLI off the ground. Applications developed using cmdkit are easy to implement, easy to maintain, and easy to understand.


An Application class provides the boilerplate for a good entry-point. Building your command-line application in layers with ApplicationGroup let’s you develop simple structures and modules that mirror your CLI.

An Interface class modifies the behavior of the standard argparse.ArgumentParser class to raise simple exceptions instead of exiting.

class Add(Application):
    """Application class for simple adding program."""

    interface = Interface('add', USAGE_TEXT, HELP_TEXT)
    interface.add_argument('-v', '--version', action='version', version='0.0.1')

    lhs: float
    rhs: float
    interface.add_argument('lhs', type=float)
    interface.add_argument('rhs', type=float)

    def run(self) -> None:
        """Business logic of the application."""
        print(self.lhs + self.rhs)

A Configuration class makes it basically a one-liner to pull in a configuration with a dictionary-like interface from a cascade of files as well as expanding environment variables into a hierarchy and merged.

The standard behavior for any good application is for a configuration to allow for system-level, user-level, and local configuration to overlap. Merging these should not clobber the same section in a lower-priority source. The Namespace class extends the behavior of a standard Python dict to have a depth-first merge for its update implementation.

config = Configuration.from_local(env=True, prefix='MYAPP', default=default, **paths)

The underlying Namespace also supports the convention of having parameters with _env and _eval automatically expanded.

password_eval = "gpg ..."

Accessing the parameter with dot-notation, i.e., config.database.password would execute "gpg ..." as a shell command and return the output.


CmdKit is tested on Python 3.7+ for Windows, macOS, and Linux, and can be installed from the Python Package Index using Pip.

$ pip install cmdkit

Getting Started#

Checkout the Tutorial for examples.

You can also checkout how CmdKit is being used by other projects.




Recommender Engine for Intelligent Transient Tracking


Hyper-shell is an elegant, cross-platform, high-performance computing utility for processing shell commands over a distributed, asynchronous queue.


A simple, cross-platform, command-line move-to-trash.

Table of Contents