ticketplease 0.1.0

CLI assistant for generating standardized task descriptions using AI

TicketPlease

CLI assistant for generating standardized task descriptions using AI.

Overview

TicketPlease is a command-line tool that helps developers and engineers generate high-quality, standardized task descriptions for platforms like Jira and GitHub using AI. It provides an interactive flow to collect key requirements and produces formatted text ready to copy and paste.

Features

• Interactive Guided Flow: Step-by-step questions to collect all necessary information, including multiline support for detailed task descriptions
• AI-Powered Content Generation: Uses LLMs to process user responses and generate complete, well-formatted descriptions
• Multi-Platform Support: Generates output in Markdown for GitHub or Jira text markup format
• Configurable Setup: Interactive wizard guides you through API key and AI model configuration when running tk config for the first time or if no configuration file exists
• Persistent Preferences: Saves user preferences for language, file paths, and platform
• Iterative Refinement: Allows users to request modifications to generated text
• Multilingual Support: Accepts input in user's language and generates output in configured language
• Clipboard Integration: Automatically copies final text to clipboard for immediate use

LLM Model Configuration

TicketPlease leverages LiteLLM to provide a flexible and configurable way to interact with various Large Language Model (LLM) providers and models. This allows you to choose your preferred AI provider (e.g., OpenAI, Anthropic, Google, OpenRouter) and the specific model you wish to use.

During the tk config setup, you will be presented with a curated list of commonly used and supported models for your chosen provider. This list is dynamically fetched via LiteLLM.

Custom Model Specification: If your desired model is supported by LiteLLM but does not appear in the default list provided during configuration, you can still specify it. Simply select the "Specify custom model" option (or similar, depending on the provider) in the wizard, and then manually enter the exact model name. TicketPlease will attempt to use this model via LiteLLM.

This flexibility ensures that you can always utilize the latest or most suitable LLM for your needs, even if it's not explicitly listed by default.

Installation

Prerequisites

• Python 3.10 or higher
• asdf for version management
• Poetry for dependency management

Installation using pip

pip install ticketplease

Development Setup

1. Clone the repository:

git clone https://github.com/kcmr/ticket-please.git
cd ticket-please

2. Install Python and Poetry using asdf:

asdf install

3. Setup the development environment:

make setup

This will install all dependencies and set up pre-commit hooks automatically.

Usage

Basic Usage

Type	Command/Option	Description
Command	tk please	Start the interactive task generation flow
Command	tk config	Configure your TicketPlease settings
Command	tk	Show help (default behavior without arguments)
Option	tk --version, -v	Show version and exit
Option	tk --help	Show this message and exit

Configuration

To configure TicketPlease, run the configuration command:

tk config

This will guide you through the setup process:

1. Choose your AI provider (OpenAI, Anthropic, Gemini, OpenRouter)
2. Enter your API key
3. Select an AI model
4. Configure default preferences

Once configured for the first time, you can update your preferences at any point by running tk config again. Additionally, many of these preferences (like output language or platform) can be overridden for individual tasks during the interactive task generation flow (tk please), providing maximum flexibility.

Configuration is stored in ~/.config/ticketplease/config.toml.

After configuration, you can start creating tasks with tk please.

Development

Available Commands

# Setup development environment
make setup

# Install pre-commit hooks
make install-hooks

# Format code
make format

# Lint code
make lint

# Lint and fix code
make lint-fix

# Run tests
make test

# Run tests with coverage
make test-cov

# Run all checks (format, lint, test)
make check

# Clean build artifacts
make clean

Pre-commit Hooks

The project uses pre-commit hooks to ensure code quality:

• Code formatting: Automatically formats code with ruff
• Linting: Checks code style and potential issues
• Commit message validation: Ensures commit messages follow conventional commits format
• File checks: Removes trailing whitespace, fixes end-of-file issues, etc.

Hooks are automatically installed when you run make setup. To manually install them:

make install-hooks

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run the development checks: make check
5. Commit your changes using conventional commits
6. Push to your fork and create a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.