01Latent criteria discovery through proactive probing and pre-mortem analysis
02Support for executable verification using Bash scripts and LLM subagents
03Automated manifest generation with structured deliverables and acceptance criteria
04Automatic integration with CLAUDE.md for project-specific standards and commands
05Configurable quality gates for code bugs, type safety, and maintainability
068 GitHub stars