What problem does it solve? Writing a new volume for the AI-ENGINEERING-OS knowledge base involves many strict contract rules — section lists per volume type, word-count minimums, executable examples with tests, and honest status tracking — that are easy to get wrong when done manually. This Skill walks through the entire production procedure with verifiable outputs at every step. ## Core Features & Use Cases - Contract-driven scaffolding: Resolves the volume's type, applicable sections, and required diagrams from contrato.json, then materializes the folder structure idempotently. - Executable examples with tests: Enforces the convention that every cited code block exists as a file under exemplos/ with a matching pytest test. - Two validation gates: Runs the structural validator (ferramentas.validar) and the executable gate (pytest), then records an honest status (RASCUNHO, never PRONTO) and logs the work in the CHANGELOG. - Use Case: A maintainer receives the request "/novo-volume 07 prompt-engine" and uses this Skill to read the contract, write all sections, create tested examples, run both gates, and register the result. ## Quick Start Ask the AI to run /novo-volume with the volume number and name, for example producing volume 07 on prompt engineering end to end.
