Data, Profit & Decision AI

xlsx

xlsx is an ecommerce AI skill for Anthropic, built for teams working with Codex, Claude Code. Use it to you need operating data to decide the next budget, product, or…

Provider
Anthropic
Platforms
Codex · Claude Code
View original link · Skills.sh

What this skill helps you do

Official Anthropic skill for working with spreadsheet files (.xlsx, .xlsm, .csv, .tsv) using pandas and openpyxl. Enforces financial model color-coding standards, formula construction rules, and automatic recalculation via LibreOffice. Scans for all Excel errors and guarantees zero formula errors on delivery. The most-installed skill on skills.sh with 136K+ installs across agentic coding platforms.

Install and get started

Copy the full instructions into your AI tool. Test one low-risk example before connecting real store data.

Original Skill instructions

You are an Excel expert. Use pandas for data analysis, openpyxl for formulas and formatting. ALWAYS use Excel formulas — never hardcode calculated values. After creating/modifying, run scripts/recalc.py to recalculate and scan for #REF!, #DIV/0!, #VALUE!, #N/A, #NAME?. Financial models: blue text for inputs, black for formulas, green for internal links, red for external links, yellow background for key assumptions. Format years as text, currency with $#,##0 and unit headers, zeros as "-", percentages as 0.0%. Place ALL assumptions in separate cells with cell references. Document hardcoded values with Source annotations. Ship with ZERO formula errors. Preserve existing template conventions.

Useful tasks

  • Creating new Excel spreadsheets from scratch with formulas, formatting, and data validation
  • Reading and analyzing spreadsheet data using pandas for statistical analysis and visualization
  • Modifying existing spreadsheets while preserving formulas, formatting, and template conventions
  • Building financial models with industry-standard color coding and number formatting
  • Converting between tabular file formats (.xlsx, .csv, .tsv) with data cleaning

How to use it

  • Always run scripts/recalc.py after creating/modifying files with formulas — openpyxl doesn't evaluate them
  • Use pandas for data analysis (df.head(), df.describe()), openpyxl for formatting and formulas
  • Never open with data_only=True and then save — formulas are permanently replaced with values
  • For large files, use read_only=True for reading or write_only=True for writing with openpyxl
  • Test 2-3 sample formula references before building the full model to catch column mapping errors early

More skills for this workflow

Content checked: