Math Modeling Skill
数学建模竞赛全流程助手:审题选题、模型选型与求解、结果检验(误差/灵敏度/稳健性)、论文写作与提交自检, 覆盖国赛 CUMCM、研赛华为杯、美赛 MCM/ICM。当用户准备建模竞赛、写建模论文、问赛题思路、写摘要、 做灵敏度分析、选模型、检查论文格式或提交前自检时使用——即使没说"数学建模"三字也应使用。
Install
npx skills add anticipate218/math-modeling-skill --skill math-modeling-skillgit clone --depth 1 https://github.com/anticipate218/math-modeling-skill.git /tmp/math-modeling-skill
mkdir -p ~/.claude/skills
cp -r /tmp/math-modeling-skill/. ~/.claude/skills/math-modeling-skillPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# 数学建模竞赛助手 帮用户把一道赛题变成**评委愿意给高分的论文**。核心不是"算出答案",而是让模型的**每一步都可解释、可复现、可检验**。 ## 三条铁律 1. **问题决定方法,不是方法决定问题。** 先读懂题目要什么,再从方法库挑;堆砌高级模型是新手最常见的失分方式(评委一眼看得出你只是把学过的模型全写上)。 2. **没有检验的结果等于没有结果。** 任何结论都要配误差、对比基线或灵敏度分析。只报"我们算出 X"的论文拿不到高分。 3. **可复现性是底线。** 论文里的每个数字都要能由附录代码重跑出来。国赛规范明确:程序不能运行或结果与论文不符,可能被取消评奖资格。 ## 第 0 步:开工前先确认三件事 问清楚(或从上下文推断)再动手,避免做错方向: | 要知道 | 为什么重要 | |---|---| | **哪个竞赛 / 什么题号** | 决定了格式硬规则、页数、语言、评审偏好 | | **还剩多少时间** | 决定策略:时间紧就主攻一问 + 保证检验与写作完整,而不是贪多 | | **交付物是什么** | 论文 PDF?支撑材料压缩包?代码?Summary Sheet? | 竞赛差异(格式、页数、语言、评审侧重)见 `references/contests.md`。 ## 第 1 步:审题与选题 1. **逐句拆题**:把题目拆成"已知条件 / 待求目标 / 隐含约束 / 数据清单"四类,写成清单。 2. **识别问题类型**:优化、预测、评价、机理、仿真、分类——多数赛题是多问递进的混合体。 3. **选题目**(若题目可挑): - 优先选**数据齐全、能算出确定结果**的题;数据缺失或需要大量外部调研的题风险高。 - 避开"看起来简单但无法定量验证"的题——评委要的是可检验的结论。 - 若队伍有专业背景优势(物理/医学/经济),优先选对口的题。 4. **给每问写一句"可验证的目标"**:例如"给出 3 个配送中心选址方案,使总成本最小且满足 24 小时内送达"。含糊的目标必然导致含糊的模型。 ## 第 2 步:问题分析 → 方法选型 **先明确每问的输出形态**(一个数值?一个方案?一个预测序列?一张评价排序表?),再选模型。 打开 `references/model-library.md` 的**快速检索表**,按"赛题信号"定位 2–3 个候选方法,然后: 1. 对每个候选写一句**为什么适用**与**为什么不选它**(论文里可以直接用这段论证)。 2. **优先选能算出、能验证的简单模型**,再用组合/改进体现创新。评委更认可"简单模型 + 严谨检验"而不是"复杂模型 + 无法验证"。 3. 多问赛题要在**同一模型体系上递进**,不要每问换一套毫不相干的模型。 4. 若问题涉及不确定性,从一开始就设计好**随机性怎么处理**(场景、区间、蒙特卡洛),不要等到最后补一句"考虑随机因素"。 ## 第 3 步:假设与符号 - **假设必须"可落地"**:每条假设后面在建模中都要真的用到,并在论文里说明它的影响。写了 10 条假设却只用 2 条,是典型的凑数。 - 假设来源:题目给定条件 + 合理的简化(忽略次要因素)+ 数据可得性限制。 - **符号说明**用表格:符号 / 含义 / 单位。全文符号唯一,正文与附录一致。 - 单位必须统一——同一模型里混用小时与分钟、元与万元,是低级但致命的错误。 ## 第 4 步:建模与求解
Details
- Skill name
math-modeling-skill- Repository
- anticipate218/math-modeling-skill
- License
- MIT
- Repo updated
- 10 October 2026
FAQ
What does the Math Modeling Skill skill do?
数学建模竞赛全流程助手:审题选题、模型选型与求解、结果检验(误差/灵敏度/稳健性)、论文写作与提交自检, 覆盖国赛 CUMCM、研赛华为杯、美赛 MCM/ICM。当用户准备建模竞赛、写建模论文、问赛题思路、写摘要、 做灵敏度分析、选模型、检查论文格式或提交前自检时使用——即使没说"数学建模"三字也应使用。
How do I install the Math Modeling Skill skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add anticipate218/math-modeling-skill --skill math-modeling-skill. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Math Modeling Skill popular?
It lives in anticipate218/math-modeling-skill, which has 55 GitHub stars and 2 forks.
Related skills
Pencilplaybook
53stevembarclay/pencilplaybook
PencilPlaybook is the UI Skills / Taste-Skill for Pencil.dev + Claude Code — a design playbook that gives Claude real perceptual psychology and senior-level guardrails so it stops producing averaged-out AI slop.
Humanize Chinese
29swaylq/humanize-chinese
中文去 AI 腔一整套(装完即用,零 API Key —— 装了本 skill 的 agent 自己就是那个 LLM)。 六个流程:① 生成即去 AI —— 写中文时按本 skill 的写作指南落笔,初稿就不带 AI 腔; ② 复查改写 —— 产出后对照本 skill 的清单再过一遍,定点拆模板句式; ③ Python 词语替换 —— ./humanize replace 按文体自动路由词表(学术 120+ 条 / 通用 220+ 条),只做短语级、只在子句边界替换,改完自动核对数字/专名/段落; ④ Claude 腔转中文 —— 把 Claude 助手腔的中文改写成平实中文, 规则提炼自 gvzdv/claudish-to-english 并本地化; ⑤ 去水印 —— ./humanize watermark 扫清零宽字符/双向控制符/同形替身, 并对照 guillaumemeyer/watermarks-remover 说明统计水印的边界; ⑥ 纠错顺句 —— 修错词、病句、标点,拗口和用词不准处轻手顺一顺;改动范围有硬校验。 另有检测(0-100 分)、风格转换、学术降重等 CLI。 Use when user says: "去AI味", "降AIGC", "人性化文本", "humanize chinese", "AI检测", "AIGC降重", "去除AI痕迹", "文本
Ue5 MCP
47ibrews/ue5-mcp
Field manual for driving Unreal Engine 5.7 / 5.8 through MCP. Engine-level gotchas, silent-fail edges, crash patterns, and the call sequences that actually work — applicable regardless of which MCP server you're using (Epic's official ModelContextProtocol plugin, custom servers, or anything else). Auto-trigger when Unreal Engine MCP tools are detected in a session, or when the user mentions Unreal Engine, UE5, Blueprints, Niagara, MetaSound, materials, or any UE editor automation workflow. This skill contains hard-won knowledge from real debugging sessions — most entries trace back to an actua
Conference Watch
67Zsun79/ConferenceWatch
Watch and report on upcoming AI/ML conferences: submission deadlines (every stage), dates, location, special sessions / calls for papers, official links, and a 5-year acceptance-rate trend. Use when the user asks about future AI conference deadlines, wants to plan submissions, compare venues, or track a set of conferences by area/reputation/difficulty. Works in any agent that has a web-search / web-fetch capability (Claude Code, Codex, etc.).
Record and Replay
48ugarchance/record-and-replay-skill
Record the user's actions — browser (Playwright event stream + trace) or native desktop (lightweight event/window/screenshot stream) — on macOS, Windows and Linux for up to 30 minutes, then turn the demonstration into a reusable agent skill. Use when the user asks you to watch them perform a task, record a workflow, or create/refine a skill from their demonstration. Agent-agnostic — works in Claude Code, Codex, opencode, or any agent with shell access.
World Model Method
49Renhuai123/world-model-method
A general-purpose working mode that makes the assistant produce output the "world-model way" — define an explicit goal and cost, build an abstract model of the situation, predict the consequences of several candidate approaches, search/optimize for the best full plan, and only then render the concrete output — instead of generating it token-by-token off the top. This operationalizes Yann LeCun's objective-driven / world-model / planning ideas (JEPA, "A Path Towards Autonomous Machine Intelligence") as a reusable prompt-level method for ANY non-trivial task: important writing, decisions, strate
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.