Excel Data Analysis Skill

Read, clean and analyze spreadsheets with pandas. Reproducible answers, not a rewritten sheet.

Try in Runcell
Made by
Runcell team
Category
Data Analysis
Format
SKILL.md
Works with
Claude Code, Codex, Cursor, Runcell (Jupyter)

Showcase

Illustrative output. The real run happens in your notebook.

Cleaning log
Header row detected at row 4
12 merged cells unpacked
Revenue: '1,240.50 €' → 1240.50 (float)
Dates: 3 formats → ISO 8601
Totals reconciled against sheet: Δ 0.00 ✓
The cleaning log records what changed and why.
DataFrame
RegionMonthRevenueOrders
EMEA2026-071,240.50318
APAC2026-07986.20244
Americas2026-071,712.00402
A clean table the next step can trust.

Description

Most Excel skills are about generating spreadsheets. This one is about analysis: load a workbook, find the real header row, fix merged cells and mixed types, and answer questions with code that can be run again.

It was written for people who receive data as spreadsheets and need numbers they can defend. The agent works in pandas, keeps every transformation in a cell, and writes results back to Excel only when you ask.

  • Loads multi-sheet workbooks and detects header rows, notes and footers
  • Normalizes dates, currencies, percentages and text-as-number columns
  • Reconciles totals against the source sheet before reporting
  • Pivots, groups and joins across sheets with pandas
  • Exports a clean CSV or a formatted .xlsx summary when needed
  • Explains each step so a colleague can follow the notebook
Editor's Recommendation

Recommended by

Runcell team

Why we built this skill

The spreadsheets people actually receive have three header rows, merged cells and a total line that does not add up. This skill exists because we kept rewriting the same cleaning code. The reconciliation step, checking totals against the source before reporting, is the part that saves the day.

What you get

  • A cleaned DataFrame with a documented cleaning log
  • Answers to your questions with the code that produced them
  • An optional summary workbook or CSV export
  • A notebook you can rerun when next month's file arrives

Use it when

  • Monthly reports that arrive as messy exports
  • Reconciling numbers between two spreadsheets
  • Turning a one-off Excel analysis into a repeatable notebook
  • Analysts who know Excel and want pandas without writing it by hand

FAQ

What does an Excel skill for Claude Code do?

It teaches the agent how to handle spreadsheets properly: open workbooks with pandas or openpyxl, find the real table inside a sheet, clean types, and check totals. Claude Code can read Excel files without a skill, but the skill makes the process consistent and the results verifiable.

Is this the same as Anthropic's xlsx skill?

No. Anthropic's xlsx skill focuses on creating and formatting workbooks. This skill focuses on analysis of existing files: cleaning, reconciling and answering questions. They can be used together.

Does it work with CSV and Google Sheets exports?

Yes. Anything pandas can read works: .xlsx, .xls, .csv and .tsv, including the exports Google Sheets produces.

Why run it in Jupyter?

A spreadsheet analysis is a sequence of transformations. In a notebook each one is a cell you can inspect, and when the next file arrives you rerun the notebook instead of redoing the work.

Related Skills

View all
Data Analysis

Exploratory Data Analysis Skill

A systematic first pass over any dataset: structure, quality, distributions, relationships, and what to do next.

Runcell teamFree
Publication-style dose-response figureVisualization

DataViz Skill

Chart choice, consistent styling and journal-ready export for matplotlib, seaborn and plotly.

Runcell teamFree
Statistics

Statistical Test Selection Skill

From question and data to the right test, with assumptions checked and results reported correctly.

Runcell teamFree

Information

Format
SKILL.md (Agent Skills)
Works with
Claude Code, Codex, Cursor, Runcell (Jupyter)
Runs in
A live Jupyter kernel in Runcell, or your terminal agent
Maintained by
Runcell team
Last updated
Sep 20, 2026
Price
Free

Run this skill in your notebook

Runcell is a Jupyter-native AI agent that executes every step in a live kernel. Results are cells you can inspect and rerun.

Try in Runcell