brainstorm

Converting natural-language requests into structured planning artifacts requires evidence-first requirement discovery that inspects the codebase and writes confirmed intent into prd.md. This approach ensures scope and risk decisions are based on repository facts rather than assumptions.

Evidence-based requirement discovery checks the codebase, docs, specs, and task history to avoid asking users to confirm facts the repository already provides. Grounding requirements in existing repository facts reduces misalignment and prevents rework during implementation.

Eliciting requirements without overwhelming users requires a decision-driven approach that asks one highest-value remaining question at a time. Each question includes a recommended answer and the trade-off, allowing users to make explicit scope and risk decisions efficiently.

Generating design.md and implement.md files alongside prd.md is necessary when task complexity demands explicit scope and risk decisions. These complex engineering tasks require execution-ready checklists and acceptance criteria to ensure proper implementation planning.

This requirement planning approach targets medium to complex engineering tasks needing explicit scope and risk decisions after task-creation consent. For simple bug fixes or minor updates, the overhead of full evidence-first discovery and multiple planning artifacts is generally unnecessary.