zhnnky329Community·28 Agent Skills Included

MathModeling-skills

Gated math modeling contest workflow from problem to paper

Guides math modeling contests from problem parsing and method screening to code, figures, and final paper. Prevents stale numbers, skipped baselines, and unverifiable claims through hard gates and frozen result snapshots. Keeps modeling judgment with the human while automating data audits, risk probes, code reviews, and three final quality audits.

452|24

↗

npx skills add zhnnky329/MathModeling-skills --all -g -y

Available:

Defines the gate-driven workflow, human decision boundaries, artifact contracts, and audit rules that the agent must follow when running each modeling skill.

All Skills in This Repository (28)

Pure Emerald Level Indicators

📦 In Repo

zhnnky329

matlab-model-code-generator

Generates executable MATLAB/.m scripts from a validated method plan and cleaned data.

robustness-checker

Run robustness, sensitivity, error, and baseline checks on modeling outputs.

model-code-analyzer

Translate validated modeling plans into language-neutral coding blueprints for Python or MATLAB.

consistency-auditor

Audit cross-media consistency across tex, code, results, data, and appendix.

model-assumptions-builder

Extract and categorize assumptions into a structured planning/model_assumptions.md document.

data-auditor-cleaner

Audit and clean raw contest data with a reproducible cleaning log.

workflow-orchestrator

Coordinates gate-driven workflows across modeling-contest stages from parsing to QA.

python-model-code-generator

Generate executable Python modeling code from validated method plans and cleaned data.

figure-table-planner

Classify figures and tables by type and map them to paper sections.

paper-polisher

Polish mathematical modeling paper drafts for grammar, clarity, and formatting compliance.

python-code-reviewer

Review Python modeling scripts against a validated method plan and expected artifacts.

matlab-code-reviewer

Validates MATLAB and Beita Tianyuan scripts for runnability and traceable outputs.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install MathModeling-skills?▼

Run `npx skills add zhnnky329/MathModeling-skills --all -g -y` in your terminal to install all 28 skills globally.

What does MathModeling-skills do?▼

It walks you through a math modeling contest step by step: parsing the problem, screening methods with risk probes, generating and reviewing Python or MATLAB code, and writing the final paper with verified numbers.

Does it choose the model for me?▼

No. It screens a main candidate, a usable baseline, and an optional fallback with evidence, but you make the final method choice and your reasoning is recorded in a decision ledger.

How does it prevent wrong numbers in the paper?▼

Every number in the paper must come from a frozen snapshot file, and changing one requires logging the change, rerunning the work, and re-freezing before submission.

Does it work with Claude Code and Codex?▼

Yes. It ships identical skill trees for both Claude Code and Codex, plus a plugin package, so the same gated workflow runs on either platform.