Agent Skills › Zafer-Liu/Data-Analysis-Agent › screening

screening

GitHub

执行单变量筛选以构建候选解释变量清单。通过检验或回归评估各变量的效应方向、量级及显著性，并报告缺失率与样本量。流程涵盖Schema确认、数据校验、分析计算及可视化，同时警示多重比较风险，避免因果推断。

skills/screening/SKILL.md Zafer-Liu/Data-Analysis-Agent

Trigger Scenarios

需要进行特征筛选寻找候选解释变量评估变量与目标的相关性

Install

npx skills add Zafer-Liu/Data-Analysis-Agent --skill screening -g -y

More Options

Use without installing

npx skills use Zafer-Liu/Data-Analysis-Agent@screening

指定 Agent (Claude Code)

npx skills add Zafer-Liu/Data-Analysis-Agent --skill screening -a claude-code -g -y

安装 repo 全部 skill

npx skills add Zafer-Liu/Data-Analysis-Agent --all -g -y

预览 repo 内 skill

npx skills add Zafer-Liu/Data-Analysis-Agent --list

SKILL.md

Frontmatter

{
    "icon": "🔎",
    "name": "screening",
    "description": "进行单变量筛选并形成候选解释变量清单（feature selection 特征筛选）",
    "allowedTools": [
        "get_schema",
        "query_data",
        "run_analysis",
        "generate_chart"
    ]
}

单变量筛选

确认目标变量与候选字段，对每个候选变量执行适当的单变量检验或回归。报告效应方向、效应量、显著性、缺失率和样本量；多重比较时提示假阳性风险，不直接宣称因果。

Tool routing

Use get_schema to identify the target variable, candidate predictors, and source table.
Use query_data to verify field names, missingness, and candidate variable types.
Use run_analysis with analysis_name="Univariate_Screening" for the screening computation.
Use generate_chart on screening rankings or effect result tables after run_analysis succeeds.