research-plan-implement

You can structure an AI-driven codebase implementation using a multi-phase workflow that covers research, planning, validation, implementation, and testing. It guides changes systematically while storing findings and validation artifacts in _thoughts directories.

Planning software development with parallel agents involves orchestrating multiple agents across research, plan, and validation phases to generate faster insights. This approach enforces structured outputs and tracks progress through validation checkpoints.

A phased workflow validates implementation plans by enforcing structured validation checkpoints between the planning and implementation stages. It generates validation artifacts stored in _thoughts folders to verify changes before actual coding begins.

Yes, you can use this structured workflow for any codebase modification that requires systematic research, planning, and testing. It is designed to guide AI-assisted software development from initial research through final verification.

Research findings and implementation plans are stored as structured artifacts in _thoughts directories within the codebase. This artifact workflow ensures that multi-phase development outputs are systematically saved and tracked.

Using a structured workflow solves fragmented software development planning by enforcing phased implementation, validation checkpoints, and parallel agent orchestration. Direct prompting lacks this systematic progress tracking and artifact generation across research and testing.