XLSX Tools

Overview

Professional Excel tools for creating, reading, and manipulating .xlsx spreadsheets.

With this skill you can:

• Create new .xlsx workbooks with full control over formatting, formulas, and data layout
• Read and extract data from existing Excel files as Markdown tables or JSON
• Modify existing spreadsheets by loading, editing, and saving them back
• Recalculate all formulas and scan for errors (#REF!, #VALUE!, #DIV/0!)
• Convert between spreadsheet formats such as CSV, XLSX, PDF, ODS, and HTML

Getting Started

Add your source data or existing spreadsheets to the project workspace, then activate the XLSX Tools skill. The agent will guide you through the process step by step.

Typical workflows

• Build a financial model: Ask the agent to create a multi-sheet workbook with assumptions, calculations, and a summary dashboard. The agent uses xlsx_build with ExcelJS to produce a professionally styled .xlsx that opens cleanly in Excel.
• Extract data for analysis: When you need to inspect or process spreadsheet contents, the agent can use xlsx_extract_data to return the data as readable Markdown tables or structured JSON arrays.
• Modify an existing spreadsheet: The agent can load an existing .xlsx file with xlsx_build (using inputPath), update specific cells, add formulas, apply formatting, and save the result to a new file.
• Verify formulas: After building a formula-heavy spreadsheet, ask the agent to run xlsx_recalc to force evaluation of all formulas and check for error values.
• Convert formats: Use xlsx_convert to move between CSV, XLSX, PDF, ODS, and HTML while preserving as much structure as possible.

Tool Reference

The XLSX Tools skill exposes four specialized tools. You do not need to remember their names—the agent will choose the right one—but this reference explains what is available.

Create and modify spreadsheets

• xlsx_build: Programmatically creates new .xlsx files or modifies existing ones using the ExcelJS JavaScript library. Best for generating structured spreadsheets (financial models, reports, data exports) with precise control over formulas, styling, number formats, and layout. When an inputPath is provided, the existing file is pre-loaded so you can modify specific cells or sheets.

Read spreadsheet data

• xlsx_extract_data: Reads an .xlsx file and returns its contents as either:

  • Markdown tables (human-readable, one table per sheet)
  • JSON arrays (programmatic access, one array of rows per sheet)

  You can filter to specific sheets by name if you only need a subset of the data.

Verify formulas

• xlsx_recalc: Recalculates all formulas in an .xlsx file using LibreOffice headless, then scans the result for error values. ExcelJS stores formulas as text without evaluating them—this tool forces a full evaluation pass to catch issues like #REF!, #VALUE!, and #DIV/0!.

Format conversion

• xlsx_convert: Uses LibreOffice in headless mode to convert spreadsheets between formats, such as:
  • .csv → .xlsx
  • .xlsx → csv
  • .xlsx → pdf
  • .xlsx → ods
  • .ods → .xlsx

Best Practices

• Never overwrite the original file: The agent always writes to a new path (for example, _updated, _final, _report) so you can compare and revert.
• Use formulas over hardcoded values: Financial models should be formula-driven so assumptions can be changed without reworking the whole spreadsheet.
• Recalculate before sharing: For important deliverables, ask the agent to run xlsx_recalc to verify all formulas compute correctly.
• Reference the API docs: The agent has access to a full ExcelJS API reference covering workbooks, worksheets, cells, formulas, styling, number formats, merging, conditional formatting, data validation, images, and print setup.