0
GitHub Stars
3
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill jjuidev/jss --skill xlsx- LICENSE.txt1.4 KB
- recalc.py6.7 KB
- SKILL.md10.1 KB
Overview
This skill creates, edits, and analyzes spreadsheet files (.xlsx, .csv, .tsv) with a focus on formulas, formatting, pivot tables, and charts. It delivers dynamic, audit-ready workbooks that recalculate correctly and contain zero Excel formula errors. The skill uses pandas for data analysis and openpyxl for precise Excel formulas and formatting, and it integrates LibreOffice recalculation for validation.
How this skill works
I read or create workbooks with openpyxl to preserve formulas and formatting, and use pandas for bulk data manipulation and visualization. All calculations are implemented as Excel formulas (not hardcoded values) so models remain dynamic. After saving, I run the provided recalc.py script (LibreOffice) to recalculate every formula and scan for Excel errors, then iterate until the workbook reports no errors.
When to use it
- Build financial models that must remain dynamic and auditable
- Convert raw CSV/TSV data into formatted, formula-driven Excel reports
- Update existing templates while preserving every formatting convention
- Create pivot tables and charts based on live workbook formulas
- Validate and fix Excel formula errors across multiple sheets
Best practices
- Never hardcode calculated results; use Excel functions and references instead
- Place all assumptions in dedicated cells and reference them with absolute addresses
- Follow color and number-format standards (e.g., blue for inputs, black for formulas)
- Format years as text, currencies with units in headers, zeros shown as '-' and negatives in parentheses
- Run recalc.py after each save and fix any reported errors before delivery
Example use cases
- Create a 5-year financial projection with assumption cells, formulas, and charts
- Load transaction CSVs, normalize data with pandas, and write a formula-driven summary sheet
- Edit a client template: match exact styles, add formulas, and preserve existing conventions
- Produce a dashboard with pivot tables and charts linked to live data ranges
- Validate a large workbook for hidden #REF! or #DIV/0! errors and correct references
FAQ
I run the recalc.py script to force LibreOffice to recalculate all formulas and return a JSON error report; I fix issues iteratively until the status is success and total_errors is zero.
Will you ever overwrite a client's template formatting?
No. When updating templates I exactly match existing format, style, and conventions and never impose standardized formatting on established templates.