Mathmodel Skill logo

Mathmodel Skill

Community
rootkiller6788
mathmodel-skill

CUMCM 国赛、MCM/ICM 美赛与电工杯数学建模竞赛的端到端协作工作流,以及一般数学建模项目。Use when a user works on a modeling competition (cumcm/mcm/icm/diangong/国赛/美赛/电工杯) or asks to run/review a modeling-competition paper from problem selection through modeling, solving, robustness, writing, compliance, and final submission review. 采用 10 阶段项目流程为脊柱、建模手/编程手/论文手三角色为阶段内执行引擎、5 个独立门禁质检,并集成双引擎论文搜索、出版级可视化、LaTeX/PDF 编译与预检、Excel/PDF 等工具。Do not trigger for generic model selection, ordinary data analysis, or non-competition paper review.

Overview

Publisherrootkiller6788
Repositorymathmodel-skill
Skill namemathmodel-skill
Stars
70
Forks
5
Bundled files
256
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 256 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by rootkiller6788 on GitHub. Read the source before you install it.

Installation

Install the Mathmodel Skill AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/rootkiller6788/mathmodel-skill/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/rootkiller6788/mathmodel-skill.git \
  .claude/skills/mathmodel-skill
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Mathmodel Skill in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Mathmodel Skill on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Mathmodel Skill is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

mathmodel-skill — 数学建模三竞赛工作流(10 阶段 × 三角色 × 5 门禁)

本 Skill 把 72–96 小时的竞赛协作组织为可恢复、可检查、可交接的流程:10 阶段项目时间线(选题→拆题→建模→求解→稳健→写作→终审)为脊柱,建模手 / 编程手 / 论文手三角色负责阶段内的实际执行,5 个独立门禁质检(M1/P1/P2/W1/W2)在每个交付物边界验收,L1–L4 反馈层驱动打分与定向精修。用户回答关键问题,agent 维护状态与脚本。

  • 10 阶段:references/stage_00..09_*.md;状态统一存 <cwd>/state/decision_log.json。
  • 三角色:references/roles/{建模手,编程手,论文手}/SKILL.md,映射见 references/角色-阶段映射.md。
  • 5 门禁:references/Subagent调度.md 定义固定质检协议。
  • 全工具:tools/{figure,paper_search,docx,xlsx,pdf,latex} + scripts/latex_check/ 预检。
  • 生成论文仅供用户参考,非可直接提交作品;结构格式以当届官方规则为准。

使用前:把 使用指南.md 复制到工作区 <cwd>/。

强制执行协议

用户点名本 Skill 或任务命中时,严格执行,不降级为建议:

  1. 首次进度更新中回显:已激活本 Skill、<cwd>、当前阶段、目标竞赛与届次、计划读取的角色和工具入口。未确认的官方规则标为待核验。
  2. 进入每个阶段前实际读取该阶段角色 SKILL.md;使用 PDF、Excel、论文搜索、DOCX 或 LaTeX 时再读取对应工具 SKILL.md。知道路径 ≠ 已执行。
  3. 严格调用本 Skill 提供的脚本和模板;已有初始化、转换、编译或校验工具时,禁止手写替代实现。
  4. 任何校验预警视为未完成;只有当届官方规则或用户明确要求允许偏离时,记录“规则来源、偏离项、理由”后继续。
  5. 环境缺少引擎、搜索源或依赖时,报告阻塞并继续完成仍可验证的部分;禁止静默换工具或用较差产物冒充完成。
  6. 按 references/Subagent调度.md 在 M1/P1/P2/W1/W2 节点立即派发独立质检,禁止等全流程结束后才首次派发;作者自检不能替代独立验收。除固定质检外不主动派发其他 Subagent。
  7. 交付前运行当前阶段的全部完成门禁;任一命令未运行、退出码非零、门禁未通过时,不得声称“已完成”。
  8. 最终回复列出实际读取的入口、实际运行的关键命令与退出码、门禁状态、核心质量指标和仍存在的阻塞;不得只说“已检查”。

三竞赛 × 三模式矩阵

Competition时长语言LaTeX 引擎规则基线经验数据
cumcm72h中文xelatex / 原创 ctexartCUMCM 202691 来源 / 59 可提取样本
mcm96hEnglishpdflatex / articleCOMAP 2027n=0,无论文分位
diangong72h中文xelatex / ctex官网 2026-03-21n=0,无论文分位
Mode反馈层用途
fastL1 单次选题试跑 / sanity check
standardL1 + L2默认主流程
championshipL1+L2+L3+L4 + red-team终稿前深度评审

模式按距 deadline 自动推荐(>60h standard;<6h 直进 Stage 9 championship)。

10 阶段索引

#阶段承担角色门禁时长关键产物
0团队启动 + 资料预扫——1h启动字段、环境、规则基线
1选题(多题对比 → 1)——2-4h选择理由、放弃项、task_type
2问题拆解建模手—2-3h子问、变量、约束、依赖图
3模型选型建模手—2-4h候选模型、证据、淘汰理由
4Foundation建模手M11h题目分析报告.md + 术语表格.md
5递归子问题循环 Q1..Qn编程手P1 → P2按题分配代码、results/、figures/、复现清单.json
6稳健性 / 灵敏度编程手—2-3h稳健区间、失败边界
7模型评价 + 推广建模手—1-2h优缺点、改进、迁移条件
8论文装配与写作论文手W1 → W212-30hpaper_workspace/*.md → LaTeX/PDF
9提交前终审 + Panel论文手—2-6h最终 PDF、支持材料、Panel 记录

角色路由(三角色执行引擎)

阶段加载入口固定交付物
2/3/4/7references/roles/建模手/SKILL.md题目分析报告.md、术语表格.md
5/6references/roles/编程手/SKILL.md.py/.m 代码、results/ 表格、figures/(原始/过程/结果三类,每类≥3张且覆盖全部子问题)、results/复现清单.json
8/9references/roles/论文手/SKILL.md默认 LaTeX/PDF(完整论文-LaTeX/ + 完整论文.pdf);用户显式要求时 Word(完整论文.docx)

默认至少 8 幅正式图且覆盖全部子问题;当届官方规则或用户要求冲突时以官方为准并记录依据。

5 门禁映射

L1 每阶段轻量打分,L2 跨阶段回检,L3 独立 Panel,L4 证据校准;5 门禁为交付物边界的独立 Subagent 验收(默认开启):

门禁触发点验收物
M1 建模终检Stage 4 结束题目分析报告 + 术语表格
P1 最小可运行Stage 5 首个 Qi 切片最小可运行结果(纵向切片)
P2 编程终检Stage 5 结束代码 + 三类图 + 复现清单
W1 证据大纲Stage 8 写正文前主张—证据映射 + 大纲
W2 论文终检Stage 8 结束完整论文(LaTeX/PDF 或 Word)

门禁回执含 6 字段(范围/输入快照/状态/证据/发现 P0-P1-P2/返工);FAIL 不可被主 Agent 改写。完整协议见 references/Subagent调度.md。

加载协议(节省 token 的关键)

只在进入阶段 N 时加载 references/stage_NN_*.md,切勿一次性全读。各阶段额外加载:

  • 每阶段开头/结尾:<cwd>/state/decision_log.json 必读/必写。
  • Stage 1:competitions/<comp>/topic_specs.json;Stage 3/5:references/model_catalog.md(候选生成)再按 references/算法索引.md 读对应 assets/*.md(算法实现深挖)。
  • Stage 0/8/9:competitions/<comp>/current_rules.md 并核对官方链接。
  • Stage 8:competitions/<comp>/{winning_patterns,phrase_bank,abstract_template,paper_skeleton}.md + cumcm 的 style_guide.md;经验锚点 empirical.json 只作参考。
  • Stage 9:anti_patterns.md + rubric_overlay.json panel personas。
  • 工具:PDF→tools/pdf,Excel→tools/xlsx,搜索→tools/paper_search,画图→tools/figure,LaTeX 编译→tools/latex + scripts/latex_check/,Word→tools/docx。

收敛准则(verdict 优先级)

block(≥1 high-severity)→ pass_early(raw_min≥9 且 weighted_mean≥9)→ pass(raw_min≥7 且 weighted_mean≥8)→ pass_with_review(Stage5 任一 Qi mark_for_review)→ refine → refine_partial(仅失败 Qi)→ carryover(iter≥3)。

weighted_mean = Σ(sᵢwᵢ)/Σwᵢ,权重来自 config/dim_weights.json[<comp>][<task_type>](clamp [0.7,1.5])。此定义在 feedback_layer1_critic.md / rubrics.md / scripts/score_artifact.py 三处必须一致。

状态持久化

  • 状态源:<cwd>/state/decision_log.json(v3.1),Codex 与 Claude Code 互通;跨会话接力。
  • 每阶段开头读 current_stage 决定恢复点,结尾写核心决策 + 5 维评分,current_stage += 1。
  • Stage 5 保存 per-Qi 状态,支持 refine_partial / pass_with_review。
  • L2 跨阶段回检读此文件主动找冲突,触发定向回滚(不重做整阶段)。
  • 环境变量:MATHMODEL_STATE_DIR / MATHMODEL_COMPETITION 可覆盖。

LaTeX 策略

  • 主编译链路:scripts/render_paper.py 把 paper_workspace/*.md 按 MATHMODEL:SECTION marker 装配进 templates/latex/<comp>/main.tex(fail-closed,占位符/缺失章节直接失败),再编译 PDF。
  • 引擎分派:中文(cumcm/diangong)固定 XeLaTeX;美赛 pdfLaTeX。禁止 CUMCM 用 latexmk -pdf(会强制 pdfLaTeX)。
  • 编译前预检:python scripts/latex_check/check_latex_env.py --contest <comp> [--use-ref-bib] [--strict-class]。
  • 提交前校验:
    • scripts/latex_check/check_latex_refs.py main.tex --bib ref.bib(\cite/\ref 完整性)
    • CUMCM:scripts/latex_check/check_latex_keywords.py main.tex(关键词须为建模术语)
    • scripts/latex_check/check_pdf.py main.pdf [--max-pages N --max-size-mb M] --identity-mode strict(页数/大小/匿名)
  • 可选官方类降级:templates/latex-alt/cumcm/(内置轻量 cumcmthesis.cls 兼容类 + ctexart fallback)与 templates/latex-alt/mcm-icm/,非默认。
  • 默认 LaTeX + PDF;用户显式要求 Word 时走 tools/docx 降级路径。
  • 主树 templates/latex/<comp>/main.tex 是 render_paper.py 的唯一装配源;tools/latex/assets/templates/ 与 templates/latex-alt/ 分别是子工具镜像与降级类,改动主树模板时须同步镜像或在提交中说明原因(有 tests/test_mirror_parity.py 守护)。

根目录契约与完成判定

  • SKILL_ROOT(本文件所在目录)只读;<cwd>(用户工作目录)只写,所有产物只能写在这里;两个根目录必须不同。
  • 声称完整完成时:当前任务涉及的独立门禁均为 PASS 且通过后产物未发生未经复验的实质变化;所有计算结论来自实际运行结果;公式/表格/图表与代码一致;引用可追溯;已按当届官方规则配置并运行对应 check 脚本;所有产物位于 <cwd>,SKILL_ROOT 未被改写。
  • 环境无 Subagent 能力时,只能报告 BLOCKED 或受限交付,不得把主 Agent 自检描述为独立通过。

用户指令快捷

  • “进入/重做 stage N”“切到 mcm/cumcm/diangong”“升级到 championship”“切到 fast”“回退到 stage M”“做 L2 回检”“看进度”。

数据来源声明

  • competitions/cumcm/:91 份来源文档、59 份成功提取文本进入观察分位;不是官方阈值。
  • competitions/mcm/、diangong/:规则基线已按官方核对;经验模式为维护者启发,empirical.json 为 n=0,不得推断数值门槛。
  • 规则会变化:current_rules.md 保存最近核对日期与官方入口,正式提交前必须以当届官方通知为准。

与外部资源的关系

核心工作流可离线运行。人工补充:国赛 personqianduixue/Math_Model、datawhalechina/intro-mathmodel、dxs.moe.gov.cn 优秀论文展廊;美赛 COMAP 官网与 MCM Tutorial;电工杯中国电机工程学会论文集。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

and 140 more files.

Frequently asked questions

What does the Mathmodel Skill AI skill do?

CUMCM 国赛、MCM/ICM 美赛与电工杯数学建模竞赛的端到端协作工作流,以及一般数学建模项目。Use when a user works on a modeling competition (cumcm/mcm/icm/diangong/国赛/美赛/电工杯) or asks to run/review a modeling-competition paper from problem selection through modeling, solving, robustness, writing, compliance, and final submission review. 采用 10 阶段项目流程为脊柱、建模手/编程手/论文手三角色为阶段内执行引擎、5 个独立门禁质检,并集成双引擎论文搜索、出版级可视化、LaTeX/PDF 编译与预检、Excel/PDF 等工具。Do not trigger for generic model selection, ordinary data analysis, or non-competition paper review.

Why use Mathmodel Skill on TypingMind?

Because you install it once and use it with any model. Mathmodel Skill is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Mathmodel Skill in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/rootkiller6788/mathmodel-skill/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Mathmodel Skill?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Mathmodel Skill?

As many as you like. As long as a model supports skills, you can use Mathmodel Skill with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Mathmodel Skill AI skill free?

Yes. It is published on GitHub by rootkiller6788 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇