math-modeling-solver

math-modeling-solver is a skill for Claude Code, Codex from Lupynow/math-modeling-skills. It costs 186 tokens per session (4,814 once invoked), scanned A, original, MIT.

A Chinese-language guide for solving mathematical modeling competition problems. It covers China’s CUMCM and America’s MCM/ICM, where teams use mathematics and code to answer real-world problem statements.

In plain words
What is it for?
Use it to analyze a new contest problem, compare models such as regression or optimization, develop algorithms and code, or connect completed modeling work to a competition paper.
Why use it?
It helps teams break a contest problem into smaller tasks and choose a suitable model, algorithm, and implementation approach. Its recommendations are reference points, not automatic answers.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to analyze a new contest problem, compare models such as…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/lupynow/math-modeling-skills/math-modeling-solver
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Lupynow/math-modeling-skills --skill math-modeling-solver
Clone the repo
git clone --depth 1 https://github.com/Lupynow/math-modeling-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for math-modeling-solver

README.md
[![agentmods](https://agentmods.dev/badge/skills/lupynow/math-modeling-skills/math-modeling-solver.svg)](https://agentmods.dev/skills/lupynow/math-modeling-skills/math-modeling-solver)
Your own site
<a href="https://agentmods.dev/skills/lupynow/math-modeling-skills/math-modeling-solver"><img src="https://agentmods.dev/badge/skills/lupynow/math-modeling-skills/math-modeling-solver.svg" alt="Measured on agentmods" height="20"></a>
Per session 186 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,814 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00186 $0.04814
Opus 5 $0.00093 $0.02407
Sonnet 5 $0.00037 $0.00963
Haiku 4.5 $0.00019 $0.00481

Measured 7d ago against content hash 1e2fbea13d4c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

math-modeling-solver scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.

The scan reads SKILL.md. This mod also ships 22 executable files (references/code-templates/python/evaluation/ahp_template.py, references/code-templates/python/evaluation/entropy_weight_template.py, references/code-templates/python/evaluation/fuzzy_eval_template.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/math-modeling-solver/SKILL.md · 326 lines

How it starts

The opening of the file, as written. The whole thing — 326 lines — stays where its author put it; the contents beside it link to each section on GitHub.

数学建模竞赛解题指导

本 skill 提供的矩阵、cookbook、playbook、代码模板,全部是知识参考而非决策指令。 对于同一道赛题,不同队伍理应有不同的建模路径。矩阵里的推荐只是技术起点—— 你的任务是结合题目具体约束、数据特征和团队判断,做出有理由的选择,而非照搬推荐。

使用流程

收到解题任务后,按以下五阶段工作流操作。

Step 0: 判断用户入口

先判断用户在哪个阶段切入:

  • 有新题目文本,从零开始 → 阶段1
  • 已有问题分析结果,需要模型推荐 → 阶段2
  • 需要先查文献再看选什么模型 → 阶段1.5
  • 已确定模型,需要算法展开和代码 → 阶段3
  • 建模已完成,需要衔接论文 → 阶段4
  • 用户直接指定了模型名(如"用 GA 求解")→ 阶段3,跳过阶段1-2
  • 用户输入匹配已知题型 → 加载对应 Playbook 获取完整解题示范(12 本 Playbook 覆盖国赛 A/B/C + 美赛 D/E/F 全部题型)

阶段识别规则

用户说 切入阶段
"这道题怎么做" + 粘贴题目 阶段1
"帮我分析这道题" 阶段1
"帮我搜一下类似问题的文献" 阶段1.5
"有没有类似题目的论文可以参考" 阶段1.5
"选什么模型好" 阶段2
"XGBoost 和随机森林怎么选" 阶段2
"帮我写 GA 代码" 阶段3
"这个公式怎么推导" 阶段3
"建模做完了,准备写论文" 阶段4
"帮我写摘要" 阶段4 → 引导切换 paper skill

阶段1:拆题分析

目标:判定每个子问题的数学本质,输出结构化的分析结果。

自动加载references/problem-decomposition.md

步骤

  1. 阅读用户提供的题目文本,提取关键信息
  2. problem-decomposition.md 的方法论,对每个子问题判定数学本质类型(共 12 种):
    • 预测/回归、分类/判别、评价/排序、优化/决策、机理/物理、聚类/分组、关联/因果、博弈/策略
    • 几何/运动学、统计推断/实验设计、网络科学/图论、生态系统/环境
  3. 明确每个子问题的:输入变量、输出目标、约束条件
  4. 分析子问题之间的数据流和递进关系
  5. 输出结构化分析结果(见下方输出格式)

输出格式

## 题目拆解

### 题目概况
- 比赛类型:[国赛/美赛]
- 题型:[A/B/C/D/E/F]
- 核心场景:[一句话概括]

### 子问题分析

#### 子问题一:[标题]
- 数学本质:[预测/评价/优化/机理/分类/...]
- 输入:[哪些变量/数据]
- 输出:[需要得到什么]
- 约束:[有哪些限制条件]
- 难点:[关键挑战]

#### 子问题二:[标题]
...

### 子问题关系
[描述数据流:问题一的输出如何成为问题二的输入]

### 整体建模流程图(文字描述)
问题一([本质类型]) → [中间结果] → 问题二([本质类型]) → [中间结果] → 问题三([本质类型])

完成后:停留,等待用户确认分析结果。确认后进入阶段1.5。


阶段1.5:文献检索

目标:用文献证据支撑模型选择——知道「别人怎么解这类题」再决定「我们怎么解」。

触发条件:阶段1完成后执行。如用户明确表示不需要文献检索(例如已自行检索、有明确模型偏好、或时间紧迫),可直接进入阶段2。

自动加载../math-modeling-paper/references/literature-review.md(跨 skill 读取,仅加载第一部分「文献检索」,不展开全文)

🚨 硬性上限(防止无限搜索,必须遵守)

  • 最多执行 5 次 WebSearch 调用(不是每个检索式都搜,是总共 5 次)
  • 找到 5-8 篇高度相关论文后立即停止,不需要穷尽所有检索式
  • T2/T3 回退仅在用户明确要求时执行,不要自动级联回退
  • 如果前 3 次搜索已经找到满意结果,直接跳到步骤 4 整理证据,不再继续搜

步骤

  1. 从阶段1第4步已生成的检索关键词中,只挑最核心的 2-3 组(不是全部),中英文各至少 1 组——英文覆盖国际期刊,中文覆盖知网/万方/维普的核心期刊
  2. 告知用户:简要列出建议的 2-3 组检索式(中英文混合即可,不需要各 3-5 组),说明信源分级。详见 literature-review.md 1.3 节
  3. 询问用户:「是否需要我代为搜索?预计 3-5 次搜索即可覆盖核心文献(中英文各半)。如你已自行检索过,可以直接告诉我找到的文献,我们跳过这步。」
    • 用户同意 → 执行步骤 4,英文用 Google Scholar / Semantic Scholar,中文直接用中文关键词 WebSearch(覆盖知网/万方/维普的公开页面),不要走 T1→T2→T3 三级级联
    • 用户拒绝或已有文献 → 跳过搜索,直接请用户提供找到的文献,进入步骤 5
  4. 执行搜索(遵守硬性上限)
    • 中英文分配:5 次搜索中,英文 2-3 次 + 中文 2-3 次(中文搜索直接在 WebSearch 中用中文关键词,如 "玻璃文物" "成分分类" 机器学习
    • 中文期刊识别:搜索结果中来自 cnki.netwanfangdata.com.cncqvip.com 的链接通常对应知网/万方/维普收录论文。优先采信标注为「核心期刊」「EI 收录」「SCI 收录」「CSSCI」「CSCD」的中文论文
    • 每次搜索后检查:是否已找到 5+ 篇相关论文(中英文合计)?是→停止搜索
    • 是否已用满 5 次 WebSearch?是→停止搜索
    • 从检索结果中提取:类似问题用了哪些方法?各方法效果对比?是否有公认基准方法?
    • 标注期刊含金量:英文优先 SCI Q1/Q2,中文优先一级学报/核心期刊(见 literature-review.md 1.5 节期刊分级),普刊/会议短文尽量不引用
  5. 输出文献证据摘要(必须标注期刊级别):

Read the full file on GitHub · 326 lines

Files

What ships with it

55 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 7d ago First seen · 326 lines · 186 tokens per session scan A 1e2fbea13d4c

Subscribe to this mod's changes

math-modeling-solver is a skill published in the GitHub repository Lupynow/math-modeling-skills (308 stars, last pushed 1mo ago), licensed MIT. It adds 186 tokens to every session and 4,814 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

mathmodel-skill

An end-to-end workflow for mathematical-modeling competitions, including CUMCM, MCM/ICM, and the Electrical Cup. It guides teams from choosing a problem through modeling, solving, checking, writing, rule compliance, and final review.

handsomeZR-netizen/mathmodel-skill · 127 tokens

ez-math-model

Use when solving CUMCM, MCM, ICM, or other mathematical modeling tasks that need contest problem intake, model selection, Python solving, figures, paper writing, quality audit, and packaged deliverables.

woodfishhhh/EZ_math_model · 48 tokens

anti-defensive-writing-en

Stops defensive writing across the entire paper lifecycle — writing, revising, cutting, and organizing experiments. Treats the paper as a press conference, not a project summary, lab log, or self-audit: identify the single most publishable strength of the work and build the most favorable, complete, and persuasive…

Adkid-Zephyr/anti-defensive-writing-Skill · 138 tokens

anti-defensive-writing

A Chinese-language writing guide for presenting a research paper around its strongest supported contribution. It treats the paper as a focused academic presentation rather than a project diary or complete lab record.

Adkid-Zephyr/anti-defensive-writing-Skill · 160 tokens

math-modeling-growth

A skill that turns the results and decisions from a real mathematical modelling project into a plain-language learning or review report. Mathematical modelling uses mathematical methods to represent and study a real situation.

mantou6666/Math-Modeling-Agent-Flow · 108 tokens

weight-loss-learn

A guided course on sustainable fat loss that adapts to your level and tracks your progress. It teaches the principles behind weight change and long-term weight maintenance.

swaylq/sijiao-skill · 111 tokens