Skip to content

davepoon/buildwithclaude / plugins/all-skills/skills/mcm-o-guide

mcm-o-guide

Verified

MCM/ICM 优秀论文写作全流程指南——基于 502 篇历年获奖论文(2004–2024)语料库的结构化分析与评委点评提炼。覆盖:摘要逐句模板、6 类题型模型选择矩阵、25 页页预算分配、图表黄金法则、灵敏度/稳健性分析范式、全文一致性检查清单。当用户准备美赛、写论文、打磨论文、查结构、选模型、做摘要、看评委标准、读优秀论文时使用。也适用于任何需要高质量数学建模论文写作的场景。

FileKindSize
GUIDE.en.mdmarkdown7.8 KB
LICENSE.txtmarkdown1.0 KB
SKILL.en.mdmarkdown24.9 KB
SKILL.mdskill md22.3 KB
playbooks/coding-playbook.mdmarkdown10.0 KB
playbooks/modeling-playbook.mdmarkdown9.8 KB
playbooks/research-playbook.mdmarkdown7.6 KB
playbooks/writing-playbook.mdmarkdown10.0 KB
references/advanced-techniques.mdmarkdown11.1 KB
references/checklist.mdmarkdown5.2 KB
references/common-errors.mdmarkdown10.0 KB
references/end-to-end-case-B.mdmarkdown6.0 KB
references/end-to-end-case-C.mdmarkdown8.5 KB
references/end-to-end-case-DEF.mdmarkdown10.1 KB
references/end-to-end-case.mdmarkdown11.9 KB
references/introduction-guide.mdmarkdown6.7 KB
references/judge-commentary.mdmarkdown8.0 KB
references/memo-letter.mdmarkdown6.3 KB
references/model-innovation.mdmarkdown5.7 KB
references/model-matrix.mdmarkdown4.6 KB
references/page-by-page.mdmarkdown12.2 KB
references/quick-reference-card.mdmarkdown3.1 KB
references/self-evaluation.mdmarkdown6.0 KB
references/sentence-patterns.mdmarkdown16.9 KB
references/strengths-weaknesses.mdmarkdown6.1 KB
references/summary-template.mdmarkdown13.7 KB
references/team-workflow.mdmarkdown7.4 KB
references/top-award-strategy.mdmarkdown9.2 KB
references/top-paper-anatomy.mdmarkdown8.3 KB
scripts/README.mdmarkdown8.0 KB
scripts/benchmark-paper.mjsscript6.7 KB
scripts/check-paper.mjsscript11.9 KB
scripts/extract-pdf.mjsscript2.5 KB
scripts/parameter-sweep.mjsscript8.6 KB
templates/README.mdmarkdown3.6 KB
templates/latex-template.texother11.4 KB
templates/markdown-template.mdmarkdown5.7 KB

scripts/benchmark-paper.mjs

View on GitHub
#!/usr/bin/env node
/**
 * benchmark-paper.mjs — MCM/ICM 论文顶尖对标工具
 *
 * 用途:读取论文文本,计算关键指标,与 O 奖/顶尖 O 奖基准对比,
 *      给出"距离顶尖还差多少"的具体建议。
 *
 * 数据基准来自 502 篇真实语料(work/metrics.json):
 *   - 普通 O 奖中位数:摘要 29 数字,8 图,2 表,2 公式
 *   - 顶尖 O 奖(前 10%):摘要 93 数字,15 图,5 表,6 公式
 *
 * 用法:
 *   node benchmark-paper.mjs <paper.txt> [--top]
 *
 *   --top 模式:按顶尖 O 奖基准评估(默认按普通 O 奖)
 *
 * 输出:
 *   - 控制台:指标对比表 + 建议
 *   - 可选:--json 输出 JSON
 *
 * 依赖:无(纯 Node.js)
 */

import fs from 'node:fs';

const args = process.argv.slice(2);
const filePath = args.find(a => !a.startsWith('--'));
const topMode = args.includes('--top');
const jsonMode = args.includes('--json');

if (!filePath) {
  console.error('Usage: node benchmark-paper.mjs <paper.txt> [--top] [--json]');
  console.error('  --top  按顶尖 O 奖(前 10%)基准评估');
  console.error('  --json 输出 JSON');
  process.exit(1);
}

const text = fs.readFileSync(filePath, 'utf-8');

// ─── 基准数据(来自 502 篇真实语料分析)───
const benchmarks = {
  normal: {
    name: '普通 O 奖(中位数)',
    summaryDigits: 29,
    figs: 8,
    tabs: 2,
    eqs: 2,
    refs: 9,
    sens: 0.64,   // 64% 有灵敏度
    strength: 0.95,
  },
  top: {
    name: '顶尖 O 奖(前 10%)',
    summaryDigits: 93,
    figs: 15,
    tabs: 5,
    eqs: 6,
    refs: 12,
    sens: 1.0,    // 100% 有灵敏度
    strength: 1.0,
  },
};

const bm = topMode ? benchmarks.top : benchmarks.normal;

// ─── 提取指标 ───

// 1. 页数
const pages = (text.match(/=== PAGE \d+ ===/g) || []).length;

// 2. 摘要数字
const summaryMatch = text.match(/Summary[\s\S]*?(?=Contents|1\s+Introduction|=== PAGE \d+ ===)/i);
const summaryText = summaryMatch ? summaryMatch[0] : '';
const summaryDigits = summaryText.match(/\d+(?:\.\d+)?%?/g)?.length || 0;

// 3. 图表数(Figure N, Table N)
const figs = new Set(text.match(/\bFigure\s*\d+/gi)?.map(m => m.match(/\d+/)[0])).size;
const tabs = new Set(text.match(/\bTable\s*\d+/gi)?.map(m => m.match(/\d+/)[0])).size;

// 4. 公式数(编号方程)
const eqs = new Set(text.match(/\(\d+\)/g)?.map(m => m.match(/\d+/)[0])).size;

// 5. 参考文献
const refMatches = text.match(/\[\d+\]/g) || [];
const refs = new Set(refMatches).size;

// 6. 灵敏度章节
const sens = /\bsensitivity\b/i.test(text) ? 1 : 0;

// 7. 优劣势章节
const strength = /\bstrengths?\b/i.test(text) && /\bweaknesses?\b/i.test(text) ? 1 : 0;

// ─── 计算得分 ───
const metrics = {
  summaryDigits, figs, tabs, eqs, refs, sens, strength, pages,
};

const comparisons = [
  { key: 'summaryDigits', label: '摘要数字', user: summaryDigits, target: bm.summaryDigits },
  { key: 'figs', label: '图表数', user: figs, target: bm.figs },
  { key: 'tabs', label: '表格数', user: tabs, target: bm.tabs },
  { key: 'eqs', label: '公式数', user: eqs, target: bm.eqs },
  { key: 'refs', label: '参考文献', user: refs, target: bm.refs },
  { key: 'sens', label: '灵敏度分析', user: sens, target: bm.sens, isBool: true },
  { key: 'strength', label: '优劣势章节', user: strength, target: bm.strength, isBool: true },
];

// 评分:每项达标得 1 分,总分 7
let score = 0;
comparisons.forEach(c => {
  if (c.isBool) {
    if (c.user >= c.target) score++;
  } else {
    const ratio = c.user / c.target;
    if (ratio >= 1) score += 1;
    else if (ratio >= 0.8) score += 0.5;
    else if (ratio >= 0.5) score += 0.25;
  }
});

// ─── 建议 ───
const suggestions = [];

if (summaryDigits < bm.summaryDigits) {
  const gap = bm.summaryDigits - summaryDigits;
  suggestions.push(`摘要数字不足:当前 ${summaryDigits} 个,目标 ${bm.summaryDigits} 个(差 ${gap} 个)。每个 task 应给出 5+ 个具体数字(参数 + 结果 + 对比 + 改进率 + 区间)。`);
}

if (figs < bm.figs) {
  const gap = bm.figs - figs;
  suggestions.push(`图表数不足:当前 ${figs} 张,目标 ${bm.figs} 张(差 ${gap} 张)。每个 task 至少 3 张图(模型示意 + 结果 + 验证)。`);
}

if (tabs < bm.tabs) {
  suggestions.push(`表格数不足:当前 ${tabs} 个,目标 ${bm.tabs} 个。参数表、结果表、对比表是必备。`);
}

if (eqs < bm.eqs) {
  suggestions.push(`公式数不足:当前 ${eqs} 个,目标 ${bm.eqs} 个。每个模型至少 3-5 个编号方程。`);
}

if (refs < bm.refs) {
  suggestions.push(`参考文献不足:当前 ${refs} 篇,目标 ${bm.refs} 篇。至少 3 篇来自 5 年内。`);
}

if (!sens) {
  suggestions.push(`缺少灵敏度分析章节(致命缺陷)。70% 的 Meritorious 论文因此降级。至少 2 种方法(参数扫描 + 情景分析)。`);
}

if (!strength) {
  suggestions.push(`缺少优劣势章节。O 奖 95% 有此章节。Strengths 3 条 + Weaknesses 3 条(带解决方向)。`);
}

// ─── 输出 ───

if (jsonMode) {
  console.log(JSON.stringify({
    file: filePath,
    benchmark: bm.name,
    metrics,
    score: score.toFixed(2),
    maxScore: 7,
    comparisons,
    suggestions,
  }, null, 2));
} else {
  const line = '='.repeat(60);
  console.log(line);
  console.log(`MCM/ICM 论文顶尖对标报告(${bm.name})`);
  console.log(line);
  console.log(`文件: ${filePath}`);
  console.log(`页数: ${pages}`);
  console.log('');
  console.log('指标对比:');
  console.log('');
  console.log('| 指标 | 当前 | 目标 | 状态 |');
  console.log('|---|---|---|---|');
  comparisons.forEach(c => {
    let status;
    if (c.isBool) {
      status = c.user >= c.target ? '✅ 达标' : '❌ 缺失';
    } else {
      const ratio = c.user / c.target;
      if (ratio >= 1) status = '✅ 超标';
      else if (ratio >= 0.8) status = '🟡 接近';
      else if (ratio >= 0.5) status = '🟠 不足';
      else status = '🔴 严重不足';
    }
    console.log(`| ${c.label} | ${c.user} | ${c.target} | ${status} |`);
  });
  console.log('');
  console.log(`评分: ${score.toFixed(2)} / 7.00`);
  console.log('');

  if (score >= 6) {
    console.log('🎯 达到顶尖 O 奖水平!');
  } else if (score >= 4) {
    console.log('✅ 达到普通 O 奖水平,仍有提升空间。');
  } else if (score >= 2) {
    console.log('🟡 接近 Meritorious 水平,需加强。');
  } else {
    console.log('🔴 建议大幅改进。');
  }

  console.log('');
  console.log('改进建议:');
  if (suggestions.length === 0) {
    console.log('  ✅ 所有指标均达标,无需改进。');
  } else {
    suggestions.forEach((s, i) => console.log(`  ${i + 1}. ${s}`));
  }
  console.log(line);
}
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall