# Decile

> 执行十分位分层分析并识别高低价值群体（segmentation 分层分析）

- Skill: `zafer-liu/decile` (Agent Skill)
- Install (CLI): `npx skillmds add zafer-liu/decile`
- Raw SKILL.md: https://api.skillmd.com/api/skills/zafer-liu/decile/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: Zafer-Liu (https://skillmd.com/u/zafer-liu)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/zafer-liu/decile

---

# 十分位分析

确认排序指标、分析实体和方向，将有效样本等频划分为十组，报告每组样本量、指标范围、核心结果及累计贡献。标记空值和并列值处理方式，并给出可执行的分层策略。

## Tool routing

1. Use `get_schema` to identify the entity key, ranking metric, target metric, and source table.
2. Use `query_data` to verify field names, null handling needs, and sample distribution.
3. Use `run_analysis` with `analysis_name="Data_Decile_Analysis"` for the decile computation.
4. Use `generate_chart` on returned decile result tables after `run_analysis` succeeds.

## Implementation reference

- Tool entry: `agent/tools/business/data.py::_tool_run_analysis`
- Analysis registry: `Function/Analyze/registry.py`
- Analysis implementation: `Function/Analyze/Data_Decile_Analysis/analyze.py`
- Chart implementation: `Function/Charts_generation/chart_generate.py`