This is a structured workflow for mathematical modeling competitions like CUMCM and MCM/ICM. It breaks the process into three sequential stages: modeling analysis (pick your math model, document formulas and references), code implementation (write Python or MATLAB solvers, generate SCI-grade visualizations with semantic color palettes), and paper writing (output a proper .docx with integrated docx skill). It comes with algorithm reference docs covering optimization, prediction, evaluation, graph theory, and ML methods, plus sample winning papers and a terminology table system to keep naming consistent across stages. The role-based structure with separate SKILL.md files for each phase keeps context loading manageable. Good for teams that want guard rails around the modeling competition grind.
npx skills add https://github.com/xiaomacoltai/math-modeling-skill --skill math-modeling