Структурирование вывода математических формул

★ 7.1 · research

formula-derivation is a Claude Code skill that structures and derives research formulas by building a coherent theory line, organizing assumptions, normalizing notation, and turning scattered equations or rough notes into a self-consistent derivation package. The skill produces exactly one of three outputs: a coherent derivation for the original target, a reframed derivation with a corrected object or revised assumptions, or a blocker report explaining why the current notes cannot yet support a valid derivation. Every non-trivial step is explicitly classified as an identity, proposition, approximation, or interpretation, and the result is written to DERIVATION_PACKAGE.md by default. It is the right choice when the derivation target is not yet fixed, the main conceptual object still needs to be chosen, or raw theory notes need to be shaped into a paper-ready formula document.