Winsorize

Documents

winsorize

Winsorize is an Agent Skill for documents published in the Zafer-Liu/Data-Analysis-Agent repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

对极端值执行缩尾处理并验证影响

View on GitHub ↗

QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

Open your project in Codex.
Copy and paste the prompt below.
Review files and risks before approving any action.

Prompt to paste

Inspect the Agent Skill at https://github.com/Zafer-Liu/Data-Analysis-Agent/blob/HEAD/skills/winsorize/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/winsorize, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

缩尾处理

识别目标数值字段和异常分布，说明上下界规则后再缩尾。不得默认覆盖原始数据；比较处理前后的分布、均值和关键指标，并记录边界与受影响行数。

Tool routing

Use get_schema to identify the target table and candidate numeric columns.
Use profile_data to quantify distribution tails before modification.
Use clean_data with the winsorize operation only after the percentile bounds are clear.
Use query_data after cleaning to verify clipped values, row counts, 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
Winsorization implementation: Function/Clean/winsorize.py