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

trimming

GitHub

对数据异常值执行截尾处理并评估偏差。通过工具路由识别表结构、分析分布、在用户确认后执行清洗，并验证样本损失与指标变化，确保数据处理的透明性与可追溯性。

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

触发场景

需要对数值型列进行异常值截尾处理要求量化删除的样本数量及潜在选择偏差

安装

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

更多选项

不安装直接使用

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

指定 Agent (Claude Code)

npx skills add Zafer-Liu/Data-Analysis-Agent --skill trimming -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": "trimming",
    "tags": [
        "mode-business",
        "mode-ml",
        "mode-empirical",
        "data-preparation"
    ],
    "description": "对异常样本执行截尾处理并评估偏差（outlier 异常值）",
    "allowedTools": [
        "get_schema",
        "profile_data",
        "clean_data",
        "query_data"
    ]
}

截尾处理

先定义异常判据和业务合理范围，量化拟删除样本及其特征。仅在用户意图明确时执行，保留原始数据和可追溯输出；处理后报告样本损失及潜在选择偏差。

Tool routing

Use get_schema to identify the target table and candidate numeric columns.
Use profile_data to quantify outliers and candidate trim boundaries before modification.
Use clean_data with the trimming operation only when the user has confirmed the rule or bounds.
Use query_data after cleaning to verify row loss, boundary effects, and key metric changes.

Implementation reference

Tool entries: agent/tools/business/data.py::_tool_profile_data, agent/tools/business/data.py::_tool_clean_data
Profiling implementation: Function/Clean/data_profile.py
Trimming implementation: Function/Clean/trimming.py