math-modeling-skill
A skill that coaches math modeling contest entrants from problem analysis to checking the paper against contest rules.
- 59 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skill consists of reference materials and a checking script and calls no external services.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/anticipate218-math-modeling-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill consists of reference materials and a checking script and calls no external services.
This is third-party code. Review the repository files before installing.
What it does
The skill lays out seven steps of work on a problem: reading the statement, choosing a model, computing, validating, writing the paper and a final self-check before submission. It knows three contests, the Chinese undergraduate one (CUMCM), the Chinese graduate one and the American MCM/ICM, as well as four official scoring dimensions and the red lines that get a paper disqualified. It bundles a `references/` catalog with a model library, algorithm notes, paper structure, scoring rubric, checklists and prompt templates, and a `scripts/check_paper.py` script for paper checks. The author stresses that the skill acts as a coach, not an automatic paper writer.
Who it is for. Students and graduate students preparing a team for math modeling contests who need a coach for methods and formatting.
Good fit when
- The team reads the problems and cannot decide which one to take.
- For an optimization problem you need a model, a baseline and a way to improve it.
- Before submission the paper must pass the contest red line checklist.
- You need guidance on justifying novelty and running a sensitivity analysis.
Not a fit when
- You want the agent to write the whole work instead of the team.
- The contest is not one of the three covered and has different formatting rules.
- The task is a plain calculation with no model or paper.
Example request
Analyze this MCM problem statement, suggest two models with a baseline and draft a validation plan against the judging criteria.Limitations
The materials follow the rules of specific years, and contest requirements, especially on AI use, change, so check the current regulations before submitting. The author marks part of the content as unofficial experience. The main language of the skill is Chinese with English sections. The `check_paper.py` script needs Python 3.9 or newer. The skill does not run computations for the team and does not guarantee a prize.
How to disable. Delete the `math-modeling-skill` directory from the agent skills folder.
FAQ
Which contests does the skill cover?
The Chinese undergraduate contest (CUMCM), the Chinese graduate contest and the international MCM/ICM. A rule comparison sits in `references/contests.md`.
How do I install it?
Clone the repository with `git clone --depth 1` and place the directory in the skills folder as `math-modeling-skill`. The repository has an `INSTALL.md` the agent reads to install the skill itself.
Does the skill write the paper for the team?
No, it guides the process, suggests structure and catches rule violations. Contest rules on AI use are collected in `references/contests.md`.
Are there ready prompts?
Yes, `references/prompt-templates.md` holds templates T0 to T8 from the start of work to submission.
How do I check the paper before submission?
Run `scripts/check_paper.py` and go through the lists in `references/checklists.md`: a general one, an AI one and separate ones for the three contests.
Discussion
No questions or reviews yet.
Related
SkillsA skill that walks a student team through a math modeling contest in ten stages, from choosing a problem to submission
A Codex skill for LLM-internship resumes: tailoring to a posting, evidence checks on each line and interview question drills.
A set of 16 skills for research work: ideation, experiments, paper writing, review, rebuttal, slides and research memory.
A catalog of 28 mental-model skills for agents: diagnosis, decisions, systems, risk, strategy and idea generation
Eleven skills for investment research, technology maps and market sizing, with the skill texts written in Chinese.
A skill that breaks reviews of a paper into concerns, plans experiments and prepares material for a reply the author finishes.
README in short
The repository packs contest knowledge into a skill with progressive disclosure: the main file gives the process and details live in references. It leans on official rules and winning paper analyses, and unofficial experience is marked.