0.2.0

Kwargify Core v0.2.0 - Initial Release! 🎉

We are thrilled to announce the first public release of Kwargify Core (v0.2.0)!

Kwargify Core is a powerful Python framework designed for building, managing, and executing complex workflow pipelines, with seamless integration capabilities for AI-powered tasks. This initial release lays the foundation for creating modular, reusable, and robust workflows.

✨ Core Features

• DAG-based Workflow Definition: Construct intricate workflows using a directed acyclic graph (DAG) structure, allowing for clear dependency management and execution flow. (src/kwargify_core/core/workflow.py)
• Modular Block System: Utilize a range of pre-built blocks for common operations or extend the system by creating your own custom blocks. Blocks encapsulate specific functionalities and are the fundamental units of work. (src/kwargify_core/core/block.py, src/kwargify_core/blocks/)
  • Built-in Blocks:
    • ReadFileBlock: Reads content from files.
    • WriteFileBlock: Writes content to files.
    • AIProcessorBlock: Processes text using AI models (e.g., GPT).
    • AIExtractorBlock: Extracts structured data from text using AI.
    • DocumentTemplateBlock: Generates documents using Jinja2-like templates.
    • JsonToStringBlock: Converts JSON data to formatted strings.
• Flexible Data Flow: Easily wire outputs from one block to the inputs of another using an intuitive mapping system, enabling dynamic data pipelines.
• Command Line Interface (CLI): A comprehensive CLI (kwargify) to run, manage, register, validate, and visualize workflows. (src/kwargify_core/cli.py)
  • Initialize new projects with kwargify init.
  • Run workflows from files or the registry.
  • List and manage registered workflows.
• Workflow Registry: Version and catalog your workflows for easy discovery, reuse, and management. (src/kwargify_core/registry.py)
• Built-in Logging: Robust SQLite-based logging captures detailed information about workflow and block execution, including inputs, outputs, status, and errors. (src/kwargify_core/logging/sqlite_logger.py)
• Resume Capability: Interrupted workflows can be resumed from the point of failure, loading outputs of successfully completed blocks from the logs.
• Retry Mechanism: Configure automatic retries for blocks to handle transient errors gracefully.
• Configuration Management: Manage project-specific settings (like database paths) via a config.toml file.

🚀 Getting Started

1. Installation:
   You can install kwargify-core using pip or add it to your project with Poetry or UV:

   pip install kwargify-core
   # or
   poetry add kwargify-core
   # or
   uv add kwargify-core

   For development, clone the repository and install with Poetry:

   git clone https://github.com/kwargify/kwargify-core.git
   cd kwargify-core
   poetry install

2. Initialize a Project:

   kwargify init

   This will guide you through setting up a config.toml file.

3. Explore Examples:
   Check out the examples/ directory to see Kwargify Core in action, including:

   • simple_workflow.py
   • contract_report_workflow_cli.py (demonstrates AI integration)

🛠️ For Developers

• Project Structure:
  • Source code: src/kwargify_core/
  • Tests: tests/ (using pytest)
  • Dependencies managed by pyproject.toml (Poetry).
• Contributing: We welcome contributions! Please see the "Contributing" section in the README.md for guidelines.

🛣️ What's Next?

This is just the beginning! We plan to:

• Expand the library of built-in blocks.
• Enhance workflow visualization capabilities.
• Add more examples and detailed documentation.
• Explore further integrations and advanced features.

We are excited to see what you build with Kwargify Core! Please report any issues or suggest features on our GitHub issue tracker.