ck:cook

End-to-end feature implementation is automated by detecting user intent and routing to workflows that manage research, planning, coding, testing, and code review using specialized sub-agents. You simply state the feature request, and the system orchestrates the entire development lifecycle through mandatory review gates.

Managing code generation and testing workflows is handled by orchestrating sub-agents across sequential phases, from initial research to finalization. The system delegates critical steps to specialized agents and enforces mandatory review gates to ensure code quality before finalizing the implementation.

Yes, feature implementation supports multiple modes including parallel, fast, no-test, and code execution modes. You can select the appropriate workflow mode based on your needs, allowing you to skip tests or run processes concurrently to accelerate development.

Workflow automation requires mandatory approval at critical review gates during interactive mode, but you can bypass these checkpoints by using auto mode. Auto mode allows the system to proceed through research, planning, implementation, testing, and code review phases autonomously without manual intervention.

Subagent delegation should be used for critical steps in feature implementation where specialized focus is needed for research, planning, coding, or testing. The system automatically delegates these critical steps to appropriate sub-agents to maintain quality across the workflow automation pipeline.

Intent detection for routing feature implementation works by analyzing the user's input request and automatically determining the best workflow mode. The system evaluates the stated feature or fix requirements and selects the appropriate execution path, whether interactive, auto, fast, parallel, or no-test mode.