Skip to content
cccskills
FreePublished on GitHub

argparse-scaffolder

Generate argparse-based Python CLI applications with subparsers, type converters, and standard library patterns. Creates lightweight Python CLIs without external dependencies.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md7.8 KB
  • README.md2.1 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Argparse Scaffolder

Generate a complete argparse CLI application with Python standard library, subparsers, and best practices.

Capabilities

  • Generate Python-based argparse CLI projects
  • Create subparser hierarchies for subcommands
  • Set up custom type converters and actions
  • Configure mutually exclusive groups
  • Implement parent parsers for shared arguments
  • Set up standard Python project structure

Usage

Invoke this skill when you need to:

  • Bootstrap a CLI using Python standard library only
  • Create a lightweight CLI without external dependencies
  • Set up subparsers for command hierarchies
  • Build CLIs that need to run on any Python installation

Inputs

ParameterTypeRequiredDescription
projectNamestringYesName of the CLI project (kebab-case)
descriptionstringYesShort description of the CLI
commandsarrayNoList of commands to scaffold
pythonVersionstringNoPython version requirement (default: ">=3.8")
useSubparsersbooleanNoUse subparsers for commands (default: true)

Command Structure

{
  "commands": [
    {
      "name": "convert",
      "description": "Convert file formats",
      "arguments": [
        { "name": "input", "help": "Input file path" },
        { "name": "output", "help": "Output file path", "nargs": "?" }
      ],
      "options": [
        { "flags": ["-f", "--format"], "choices": ["json", "yaml", "xml"] },
        { "flags": ["-v", "--verbose"], "action": "store_true" }
      ]
    }
  ]
}

Output Structure

<projectName>/
├── pyproject.toml
├── README.md
├── .gitignore
├── src/
│   └── <package_name>/
│       ├── __init__.py
│       ├── __main__.py        # Entry point
│       ├── cli.py             # Argparse setup
│       ├── commands/
│       │   ├── __init__.py
│       │   └── <command>.py   # Command handlers
│       ├── types/
│       │   ├── __init__.py
│       │   └── converters.py  # Custom type converters
│       └── utils/
│           ├── __init__.py
│           └── helpers.py     # Helper functions
└── tests/
    ├── __init__.py
    └── test_<command>.py

Generated Code Patterns

Main CLI Entry (src/<package>/cli.py)

import argparse
import sys
from typing import Optional, Sequence

from .commands import convert, validate


def create_parser() -> argparse.ArgumentParser:
    """Create the argument parser with all subcommands."""
    parser = argparse.ArgumentParser(
        prog='<projectName>',
        description='<description>',
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog='''
Examples:
  %(prog)s convert input.json output.yaml
  %(prog)s validate config.yaml --strict
        '''
    )

    parser.add_argument(
        '-v', '--verbose',
        action='store_true',
        help='Enable verbose output'
    )
    parser.add_argument(
        '--version',
        action='version',
        version='%(prog)s 1.0.0'
    )

    # Create subparsers
    subparsers = parser.add_subparsers(
        dest='command',
        title='commands',
        description='Available commands',
        help='Command to run'
    )

    # Register commands
    convert.register_parser(subparsers)
    validate.register_parser(subparsers)

    return parser


def main(argv: Optional[Sequence[str]] = None) -> int:
    """Main entry point."""
    parser = create_parser()
    args = parser.parse_args(argv)

    if args.command is None:
        parser.print_help()
        return 1

    # Execute command
    return args.func(args)


if __name__ == '__main__':
    sys.exit(main())

Command Template (src/<package>/commands/convert.py)

import argparse
from pathlib import Path
from typing import TYPE_CHECKING

if TYPE_CHECKING:
    from argparse import _SubParsersAction


def register_parser(subparsers: '_SubParsersAction[argparse.ArgumentParser]') -> None:
    """Register the convert subparser."""
    parser = subparsers.add_parser(
        'convert',
        help='Convert file formats',
        description='Convert files between different formats.',
        formatter_class=argparse.RawDescriptionHelpFormatter,
    )

    parser.add_argument(
        'input',
        type=Path,
        help='Input file path'
    )
    parser.add_argument(
        'output',
        type=Path,
        nargs='?',
        help='Output file path (default: stdout)'
    )
    parser.add_argument(
        '-f', '--format',
        choices=['json', 'yaml', 'xml'],
        default='json',
        help='Output format (default: %(default)s)'
    )

    # Mutually exclusive group
    group = parser.add_mutually_exclusive_group()
    group.add_argument('--compact', action='store_true', help='Compact output')
    group.add_argument('--pretty', action='store_true', help='Pretty print')

    parser.set_defaults(func=execute)


def execute(args: argparse.Namespace) -> int:
    """Execute the convert command."""
    print(f"Converting {args.input} to {args.format}")

    if args.output:
        print(f"Output: {args.output}")

    return 0

Custom Type Converter

import argparse
from pathlib import Path


def existing_file(value: str) -> Path:
    """Type converter for existing file paths."""
    path = Path(value)
    if not path.exists():
        raise argparse.ArgumentTypeError(f"File not found: {value}")
    if not path.is_file():
        raise argparse.ArgumentTypeError(f"Not a file: {value}")
    return path


def port_number(value: str) -> int:
    """Type converter for port numbers."""
    try:
        port = int(value)
    except ValueError:
        raise argparse.ArgumentTypeError(f"Invalid port number: {value}")

    if not 1 <= port <= 65535:
        raise argparse.ArgumentTypeError(f"Port must be 1-65535: {value}")

    return port

Dependencies

[project]
name = "<projectName>"
version = "1.0.0"
requires-python = ">=3.8"
dependencies = []  # No external dependencies!

[project.optional-dependencies]
dev = [
    "pytest>=7.0.0",
    "pytest-cov>=4.0.0",
    "mypy>=1.0.0",
]

[project.scripts]
<projectName> = "<package_name>.cli:main"

Workflow

  1. Validate inputs - Check project name, commands structure
  2. Create directory structure - Set up Python project layout
  3. Generate pyproject.toml - Configure project metadata
  4. Create CLI entry point - Argparse setup with subparsers
  5. Generate commands - Individual command handlers
  6. Create type converters - Custom argument types
  7. Set up tests - pytest test structure
  8. Generate main.py - Package execution support

Best Practices Applied

  • Zero external dependencies for core CLI
  • Type hints throughout
  • Custom type converters for validation
  • Mutually exclusive argument groups
  • Parent parsers for shared arguments
  • Proper help text and examples

References

Target Processes

  • cli-application-bootstrap
  • argument-parser-setup
  • cli-command-structure-design

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Provide implementations of advanced data structures

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

Fresh adversarial code review with binary PASS/FAIL verdicts, evidence citations, and anchoring bias prevention via fresh reviewer spawning.

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

WASM-based instant code transforms for simple tasks, achieving 352x speedup over LLM inference with zero cost.

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

Coordinate Crew (persistent) and Polecat (transient) agents using Gas Town's hook-based work distribution and GUPP principle.

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

Generate visual representations of algorithm execution

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

Hierarchical coordination and drift detection with frequent checkpoints, shared memory coherence validation, role specialization enforcement, and short task cycles.

The English summary is being prepared. Showing the original description.

a5c-ai/babysitter1.9KUpdated Sep 17, 2026

See all skills by a5c-ai

Report a problem with this skill