What problem does it solve? Turning a loose feature idea into a structured requirements document is slow and error-prone, especially when the new feature must align with an existing legacy system. This Skill converts a free-form natural language description into a complete requirements.md file grounded in artifacts extracted from the legacy codebase. ## Core Features & Use Cases - Requirements Generation: Transforms a free-text feature idea into a structured requirements.md following a project template, with confidence markers and up to three [DÚVIDA] (doubt) markers for ambiguous points. - Legacy Context Anchoring: Reads reverse-engineering artifacts (architecture.md, domain.md, inventory.md, code-analysis.md, addenda) so every requirement cites its source in the legacy system. - Feature State Management: Detects in-progress features via physical artifacts, supports pausing or abandoning previous features, and manages active-requirements.json atomically. - Use Case: A developer types a one-paragraph idea for a new reporting feature; the Skill reads the legacy extraction, writes a cited requirements.md into a new feature folder, and suggests the next pipeline step (clarify or plan). ## Quick Start Ask the agent to run the reversa-requirements skill with a short description of the feature you want, for example: create requirements for adding CSV export to the orders report.
