This project is an AI agent orchestration framework that installs a complete, spec-driven software development lifecycle across multiple AI coding agents. It coordinates agents through sequential stages from discovery and requirements through design to autonomous implementation, with each task receiving independent review.
The system distinguishes itself by spawning a fresh implementer and reviewer per task, running test-driven development behind feature flags, and automatically debugging on failure. New code is introduced behind feature flags to isolate changes and enable safe rollback, while a file-based configuration system lets teams customize workflow templates and AI generation rules to match their processes.
The framework manages lifecycle stage transitions with defined entry and exit criteria, and generates structured project documents by populating customizable templates with AI-generated content.