handsomeZR-netizen

handsomeZR-netizen / mathmodel-skill

Public

三竞赛 (CUMCM/MCM/电工杯) 数学建模 skill — harness-agnostic, 同时支持 Claude Code 与 Codex CLI, 全程问答式 (Friendly Mode), 10 阶段 + 4 反馈层 + per-Qi 加权聚合 + 题型 dim 加权 + empirical 实测分位锚定

agents-md claude-code claude-skill codex-cli cumcm
15
0
100% credibility
Found May 13, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A question-driven AI workflow that guides teams through writing math modeling competition papers in 10 stages, drawing from patterns in past winning entries.

How It Works

1
📚 Discover the Competition Guide

You hear about a friendly guide that helps teams write winning papers for math modeling contests like national or international ones.

2
Pick Your Contest
🇨🇳
National Contest (CUMCM)

For the 3-day Chinese university math modeling challenge with Chinese papers.

🇺🇸
International Contest (MCM)

For the 4-day English math modeling contest with detailed problem sets.

🔌
Engineering Cup (Diangong)

For the engineering-focused Chinese contest with many sub-problems.

3
❓ Answer Easy Questions

Just reply with numbers to simple choices about your team, problem, and deadline—no typing long commands.

4
🤖 AI Builds Your Paper Step by Step

The smart assistant walks you through 10 clear stages, from planning to writing, using tips from past winners.

5
✅ Review and Improve with Feedback

Get checks at each step to fix issues early, keeping everything on track without starting over.

6
📄 Create Your Final Paper

Watch your full paper come together, complete with charts, math, and polish, ready in a printable format.

🎉 Submit with Confidence

Your competition paper is complete, structured like award-winners, and you're set to compete!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mathmodel-skill?

This Python skill automates the 10-stage workflow for mathmodel competitions like CUMCM, MCM, and Diangong Cup, turning chaotic 72-96 hour paper sprints into structured Q&A sessions via Friendly Mode. Users answer numbered prompts—no bash, Python, or JSON editing required—and get code starters, sensitivity tables, and LaTeX papers. It's harness-agnostic, working seamlessly with Claude Code or Codex CLI, with state files shared across tools.

Why is it gaining traction?

Standout features include per-Qi weighted aggregation, dim-specific scoring, and empirical percentile anchoring from 91 real CUMCM papers, ensuring outputs match winning patterns without guesswork. Fast/standard/championship modes scale effort from 30min sanity checks to 12h polishes, and CLI integration lets teams switch tools mid-competition without losing progress. Developers love the zero-friction prompts with "let AI decide" fallbacks.

Who should use this?

Math modeling undergrad teams prepping for CUMCM national rounds, especially those splitting modeling/programming/writing roles. Python users handling optimization, prediction, or simulation subtasks in 3-5 question problems. Teams mixing Claude and Codex CLI for Day 1 kickoff to Day 3 PDF rendering.

Verdict

Worth a test run for competition teams—solid docs and Python code starters make it immediately usable despite 15 stars and 1.0% credibility score. Still early (MCM/Diangong in seed mode), so expect tweaks, but it nails rhythm control better than ad-hoc LLM prompting.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.