Any number of messy spreadsheets become one clean table with columns lined up, duplicates removed by key, dates and amounts in one format, and a report of every change.
Who it is for: Operations, admin or finance staff and analysts who receive the same report as many spreadsheets (per branch, month or person) and must combine them into one table.
Format: Claude skill pack · Version: 1.0.0 · Tested on: Python 3.13.5 (standard library only)
What it does
For each file (and each sheet with --all-sheets) it finds the header row under any title lines. Column names are compared after ignoring case, spaces and punctuation; your column map renames first, and --fuzzy lines up look-alike names and lists each one. Cells are trimmed; blank, total, subtotal and repeated header rows are dropped; exact duplicates are kept once. With --key, a later row with the same key replaces the earlier one and the replaced row goes to duplicates.csv. Dates and amounts you name are rewritten in one format; cells it cannot read are listed.
What you need
- Claude.ai with code execution switched on (or Claude Code, or just Python 3.10+)
- The spreadsheets you want to combine
5-minute setup
- Claude.ai: open Settings → Capabilities → Skills, click Upload skill and pick
claude-ai-upload/sheet-merger-cleaner-pro.zip. - Claude Code: copy the folder
skills/sheet-merger-cleaner-prointo~/.claude/skills/. - Start a new chat. Attach your file and say: "Use sheet-merger-cleaner-pro. Merge every sheet in these workbooks into one table, keep the latest row per Booking ID and fix the dates".
- Claude runs the script and gives you the files listed under Outputs. Check them before you act on them.
Try it with the sample
Run the sample (three fictional regional booking exports):
python3 skills/sheet-merger-cleaner-pro/scripts/core.py samples/exports --key "Booking ID" --dates Check-in --amounts Amount --out out
You should see 9 rows in and 6 out (1 exact duplicate and 2 bookings listed in two regions), 6 columns, dates as YYYY-MM-DD and amounts as plain numbers.
Settings you can change
| Option | Default | What it is |
|---|---|---|
inputs |
(required) |
CSV/XLSX files or a folder |
--all-sheets |
off |
Merge every sheet of each workbook |
--map |
(none) |
Column map: one 'old name = new name' per line |
--fuzzy |
off |
Line up look-alike column names |
--key |
(none) |
Column(s) identifying a row; the later row wins |
--dates / --amounts |
(none) |
Columns to rewrite as YYYY-MM-DD or plain numbers |
--us-dates |
off |
Read 03/04/2026 as March 4 |
--out |
out |
Output folder |
What goes in and what comes out
Inputs: Any number of CSV, TSV or XLSX files (or a folder); Optional: column map, key column, date and amount columns
Outputs: merged.csv with a source file (and sheet) column; report.md with every change counted; duplicates.csv; files.csv
AI disclosure and limits
This product does not call an AI model at run time. It was built by Oarsman AI's AI agents and checked by the automated tests shipped with it.
Limits: - Reads .csv, .tsv and .xlsx; save .xls as .xlsx first. - Formulas are read as their stored values. - Look-alike matching is for names that differ by a letter or two; very different names need the column map.
Not tested by us: - .xls (old Excel) files - merged cells and formulas (values are read as stored) - XLSX inputs in the shipped tests (CSV only)
Files
- oarsman-sheet-merger-cleaner-pro-v1.0.0.zip
- CHANGELOG.md
- QA-REPORT.json
- README.md
- SPEC.md
- SUPPORT.md
- TROUBLESHOOTING.md
- VERSION
- VIDEO-SCRIPT.md
- claude-ai-upload
- diagram.mmd
- diagram.png
- preview.png
- samples
- skills
- tests
Help
Self-serve: see TROUBLESHOOTING.md (fixes and FAQ). There is no call, chat or email support. Fixes ship as new versions at your download link. Refund: 14 days, works as described, through the checkout platform.