kli-workflow

AI development lifecycle orchestration coordinates transitions between research, planning, implementation, and reflection phases while automatically loading phase-specific skills. It streamlines the workflow by ensuring structured progression through each development stage.

Observation-driven learning captures knowledge through an event stream during task execution, facilitating continuous playbook evolution. It processes event streams to improve development workflows based on real implementation observations.

Task graph management and event stream processing require a CLI and MCP server to coordinate the development lifecycle. These dependencies enable phase transitions, skill loading, and observation-driven learning loop functionality.

Starting the research phase for a new feature begins the lifecycle orchestration process, which then guides progression through planning, implementation, and reflection. The workflow automatically loads phase-specific skills for detailed guidance at each stage.

Managing transitions between research and implementation phases uses lifecycle orchestration to coordinate stage changes and load appropriate phase-specific skills. This ensures all development steps are covered and learnings are captured for continuous improvement.

Task management orchestration captures knowledge and evolves playbooks automatically through an observation-driven learning loop. It leverages event stream processing to facilitate continuous improvement of development playbooks across lifecycle phases.