What problem does it solve? It turns a structured implementation plan (actions.md) into executed code changes with full traceability, recording progress, legacy impact, and regression watch items so nothing is lost between planning and auditing. ## Core Features & Use Cases - Phased plan execution: Runs pending actions phase by phase (Preparação, Testes, Núcleo, Integração, Polimento), respecting dependencies and parallel markers, and updates checkboxes from [ ] to [X]. - Audit trail generation: Produces legacy-impact.md (mapping touched files to architecture components with severity) and regression-watch.md (stable-ID watch items for modified or removed rules). - Dual context anchoring: Works anchored on legacy extraction artifacts (architecture.md + domain.md) or greenfield artifacts (prd.md + SDD specs), aborting cleanly when neither exists. - Use Case: After running /reversa-to-do to build an action plan for a feature, invoke this skill to execute the plan, then hand legacy-impact.md and regression-watch.md to a future /reversa re-extraction cycle. ## Quick Start Type "/reversa-coding" or ask the agent to execute the active feature plan so it codes the pending actions and writes the progress and audit files.
