A structured, three-stage workflow skill designed for mathematical modeling competitions (CUMCM, MCM/ICM). It provides a comprehensive system encompassing modeling analysis, code implementation, and paper writing, supported by an extensive algorithm library and advanced visualization tools, aiming to enhance the quality and efficiency of competition submissions.

How It Works

This skill employs a collaborative, role-based approach with three distinct phases: "Modeling Hand" for analysis and design, "Coding Hand" for implementation and visualization, and "Writing Hand" for argumentation and paper composition. It leverages a rich algorithm resource library (60+ algorithms across 7 categories) and emphasizes SCI/Nature-grade visualization through a "Figure Contract" planning system, semantic color palettes, and dual SVG/PNG output. Writing is argument-driven, utilizing "Claim-Evidence" mapping and structured templates, integrated with document processing and academic search tools.

Quick Start & Requirements

• Installation:
  • Git clone: git clone https://github.com/XiaoMaColtAI/math-modeling-skill.git and place in the agent's skills directory.
  • npx skills add https://github.com/xiaomacoltai/math-modeling-skill --skill math-modeling
  • Download ZIP.
• Recommended Agents: Claude Code (preferred), Codex, Cursor, Trae, Qoder, Hermes Agent, OpenClaw.
• Prerequisites: Standard AI agent environment. The paper_search skill requires an email for OpenAlex and optionally an API key for AnySearch.
• Links: OpenAlex API, OpenAlex Docs, AnySearch API Keys.

Highlighted Details

• Three-Stage Collaboration: Distinct roles (Modeling Hand, Coding Hand, Writing Hand) for modeling analysis, code implementation, and paper writing.
• Extensive Algorithm Library: Over 60 algorithms across 7 categories (Optimization, Prediction, Evaluation, Graph Theory, Statistics, Comprehensive, Machine Learning).
• SCI/Nature-Grade Visualization: Pre-planned "Figure Contract," semantic color palettes, SVG/PNG dual format, and HTML navigation panels.
• Argument-Driven Writing: "Claim-Evidence" mapping, chapter architecture patterns, self-review framework, and English workflow support for US competitions.
• Integrated Document Tools: Specialized sub-skills for docx, pdf, xlsx processing, and paper_search for academic literature.
• Active Development: Major updates in v2.4 (May 2026) focusing on modularization, role-specific enhancements, and visualization/search improvements.

Maintenance & Community

The project shows active development with significant updates in May 2026. No explicit community links (Discord, Slack, etc.) or contributor/sponsorship details are provided in the README.

Licensing & Compatibility

• License: MIT License.
• Compatibility: The MIT license permits broad use, including commercial applications and linking with closed-source projects, provided attribution is maintained.

Limitations & Caveats

AI-generated papers may be detectable; users are strongly advised to use generated content as a reference and revise it significantly with personal insights and style adjustments. The paper_search skill requires configuration (email for OpenAlex, optional API key for AnySearch).