Agent skill · Data & Analytics

mcm-c-coder

Coding specialist for COMAP MCM/ICM Problem C (C题数据处理与建模落地). Use when you need to implement reproducible data pipelines, feature engineering, model training/validation, bootstrap intervals, and figure/table generation.

majiayu000github.com/majiayu000GitHub ↗
claude-codeMIT
Install
npx skills add majiayu000/claude-skill-registry --skill mcm-c-coder --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 3 KB
Bundled scripts: none
Path: skills/ai-ml/mcm-c-coder/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# MCM/ICM C题代码手(数据处理与建模落地) ## 目标 - 把题面附件数据变成“可复现的分析流水线”(读取→清洗→特征→训练→评估→出图/表)。 - 确保:无泄露、可重复、可解释、输出能直接用于论文。 ## 默认技术栈(可根据项目约束调整) - 包管理:`uv`(默认) - Python:`pandas/numpy/scipy` - 建模:`scikit-learn`(通用),必要时 `statsmodels`(统计推断) - 可视化:`matplotlib/seaborn` ## 默认项目骨架(建议) 目标:让“拿到题面数据 → 一键复现图表/表格/区间”的路径最短。 - `data/raw/`:题面原始附件(只读) - `data/processed/`:清洗后的数据(可复现生成) - `src/`:特征、训练、评估核心逻辑 - `scripts/`:可执行入口(例如 `scripts/run_all.py`) - `outputs/figures/`:论文用图(png/pdf) - `outputs/tables/`:论文用表(csv/tex) - `outputs/predictions/`:预测与区间结果(csv) ## uv 依赖管理(默认约定) - 使用 `pyproject.toml` 管理依赖,`uv.lock` 锁定版本。 - 依赖分组建议: - **runtime**:`pandas numpy scipy scikit-learn matplotlib seaborn` - **dev**:`ruff black`(可选) 可复现要求:报告中出现的任何图表/表格,必须能由 `uv` 环境运行脚本生成。 ## 工作流 1. **建立数据入口(I/O 层)** - 读取 csv/tsv/xlsx 时固定:编码、缺失值标记、日期解析、dtype。 - 对每个原始文件输出:行数、列数、主键唯一性、时间范围。 2. **数据清洗(Clean)** - 统一:列名风格、单位、时区/日期格式。 - 处理: - 重复(按主键去重或聚合)。 - 缺失(删除/插补/单独类别/用模型可接受的方式)。 - 异常(winsorize/截断/规则过滤)。 3. **特征工程(Features)** - 时间序列/事件序列:滚动窗口、滞后项、变化率、累计量。 - 类别:one-hot/target encoding(注意泄露)。 - 文本(如 2020C):TF-IDF/情感词典/简易主题(优先轻量可解释)。 4. **切分与验证(No Leakage)** - 时间序列:滚动窗口验证(不要随机打乱)。 - 结构化独立样本:train/valid/test + CV。 - 任何“当天决策”题(如交易策略):严格使用 `<= t` 数据。 5. **训练与

What's inside
Steps it walks through
  1. 目标
  2. 默认技术栈(可根据项目约束调整)
  3. 默认项目骨架(建议)
  4. uv 依赖管理(默认约定)
  5. 工作流
  6. 质量门槛(交付前自检)
Ships with 1 file
  • metadata.json
More from claude-skill-registry
All skills →
About this skill
What does the mcm-c-coder skill do?

Coding specialist for COMAP MCM/ICM Problem C (C题数据处理与建模落地). Use when you need to implement reproducible data pipelines, feature engineering, model training/validation, bootstrap intervals, and figure/table generation.

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill mcm-c-coder --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going