xlsx — 表格数据读写

SkillFiles & storage

Use when EZ_math_model needs to read, preview, clean, write, or summarize CSV, XLS, or XLSX attachments and result tables for modeling tasks.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the xlsx — 表格数据读写 skill

What this skill tells your AI

The instructions your AI receives, as published by woodfishhhh/ez_math_model in skills/ez-math-model/tools/xlsx/SKILL.md and read by ahel’s review.

何时使用

  • 读取 .xlsx.xls.csv 附件。
  • 写出 results/qN_summary.csv.xlsx
  • 生成 intake.json.attachments[i].preview
  • 大 CSV 需要分块处理。

预览契约

输出到 intake preview:

{
  "shape": [1024, 8],
  "columns": ["timestamp", "value"],
  "head_3": [{"timestamp": "2024-01-01", "value": 12.3}],
  "missing_per_col": {"value": 5}
}

优先链路

  1. pandas 读写和清洗。
  2. openpyxl 写公式或保留 Excel 格式。
  3. 宿主 xlsx skill 处理公式刷新、LibreOffice 重算等高级场景。

pandas 模板

import pandas as pd

df = pd.read_excel("attachments/data.xlsx")
preview = {
    "shape": list(df.shape),
    "columns": df.columns.tolist(),
    "head_3": df.head(3).to_dict("records"),
    "missing_per_col": {c: int(n) for c, n in df.isna().sum().items() if n > 0},
}
df.to_csv("results/q1_summary.csv", index=False, encoding="utf-8")

CSV 编码尝试顺序:utf-8gbkgb2312latin-1

大文件协议

  • pd.read_csv(path, chunksize=200_000)
  • 提前指定 dtype
  • 必要时抽样 5% 做 EDA。
  • 处理完及时释放分块 DataFrame。

失败诊断

情况处理
编码全失败写诊断,建议用户提供 utf-8 版
列名含非法字符生成 ASCII 安全别名并保留 column_alias.json
文件超过 1GBchunksize 分块处理

Signals

GitHub stars
41
Forks
1
Last commit
Jul 2026
Hacker News mentions
10
Advanced
Catalog kind
skill
Gateway key
xlsx-woodfishhhh
Source
github.com/woodfishhhh/ez_math_model