Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Math Modeling Contest Skill

Runs a full math-modeling contest workflow across three roles — modeler, coder, paper writer — with independent QA gates at each stage.

EducationAdvanced★ 661⑂ 34AI score 8/10Last updated: Aug 10, 2026

What it does

Splits math-modeling contest work into three separate roles with enforced completion gates:

  • Modeler: reads the problem statement and attachments, produces a problem-analysis report and a notation/terminology table, and picks at most two independent model systems per sub-question.
  • Coder: writes runnable Python (.py) or MATLAB (.m) code, exports result tables (.csv/.xlsx), generates at least 9 figures (raw-data / process / result, ≥3 each and covering every sub-question), plus results/复现清单.json recording random seeds, input SHA-256 hashes, dependency versions and a single reproduction command.
  • Writer: outputs 完整论文.docx (Word) by default, and on request a LaTeX project with a genuinely compiled PDF — checked for native OMML equations, figure/citation numbering, page counts, embedded fonts and image DPI.
  • Independent review: at checkpoints M1/P1/P2/W1/W2 a separate read-only subagent that did not author the artifact verifies it; a FAIL sends work back to the responsible role.
  • Enforces a strict SKILL_ROOT (read-only) vs PROJECT_ROOT (all new output) separation so templates and references are never overwritten.

Who it's for

  • Students and teams preparing for math-modeling competitions (CUMCM, APMCM, MCM/ICM style)
  • Researchers who need a chain from model design → runnable code → publishable draft
  • Anyone who wants reproducible, evidence-backed results rather than invented numbers
  • Note: the skill's internal documentation is written in Chinese

Usage examples

  1. End-to-end: "Here's the contest problem PDF and the data workbook — do the modeling, the code and a Word paper." → runs the three stages in order and reports each gate's PASS/FAIL status.
  2. Modeling only: "Just analyze the problem and recommend models per sub-question." → loads only the modeler role and produces the analysis report and notation table.
  3. Code and figures only: "I already have the model spec — implement it in Python and produce raw/process/result plots for every sub-question." → runnable scripts, ≥9 figures, and a reproduction manifest JSON.
  4. Add LaTeX output: "I also need a LaTeX version." → generates the LaTeX source project, actually compiles the PDF, and writes hash-bound build metadata.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/XiaoMaColtAI/math-modeling-skill/HEAD/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the SKILL.m folder from the GitHub repo XiaoMaColtAI/math-modeling-skill into my ~/.claude/skills/math-modeling/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/XiaoMaColtAI/math-modeling-skill.git ~/.claude/skills/math-modeling

⚠ This is a third-party skill. Check the source repository before installing.

  1. Make sure Claude Code is installed and you can use a terminal.
  2. Create the skills directory: mkdir -p ~/.claude/skills
  3. Clone the repo: git clone https://github.com/XiaoMaColtAI/math-modeling-skill.git ~/.claude/skills/math-modeling
  4. Verify that ~/.claude/skills/math-modeling/SKILL.md exists (if it sits in a subfolder, move it to the right level).
  5. Recommended prerequisites: Python with numpy, pandas, matplotlib and scipy for the coding stage; TeX Live and pandoc for the paper stage. Missing tooling makes the skill report BLOCKED instead of finishing.
  6. Create a separate working folder (your PROJECT_ROOT) holding the problem files and data, then start Claude Code inside it.
  7. Ask something like "use the math modeling skill to solve this problem"; the first reply should echo SKILL_ROOT, PROJECT_ROOT, the target contest and the stage plan.
  8. Always confirm formatting against the current year's official contest rules and template — the generated paper is a reference draft, not a submission-ready deliverable.
View source on GitHub ↗