
XLSX
Excel workbooks with live formulas: build or edit .xlsx models with openpyxl...
Install
npx skills add ginlix-ai/LangAlpha --skill xlsx -g -a claude-code -y- Repository
- ginlix-ai/LangAlpha2.2k
- Category
- Documents
- Updated
- 3d ago
- License
- Apache-2.0
- Context cost
- Mediumabout 3,000 tokens
- Runs scripts
- Yes, review before use
What it does
Excel workbooks with live formulas: build or edit .xlsx models with openpyxl, recalculate with IronCalc or LibreOffice, audit conventions, profile messy uploads, render for review
XLSX
Build or edit an Excel workbook and write it into the task directory (e.g. dcf_acme/acme_dcf.xlsx). The user opens it in Excel, changes an input, and every dependent number moves. That is the whole point of the format: a workbook the agent delivers is a model, not a table of results. A cell that could be computed from other cells is a formula. A number typed into a formula's place is a defect, even when the value is right today.
This is the right output when the user wants something they will keep working in: a DCF, a comps sheet, a three-statement model, a screen they will re-sort, a schedule they will extend. It is the wrong output for a one-off table (put it in the chat or an HTML report) and for a document that will not be edited (see .agents/skills/html-report/SKILL.md).
User preferences override these defaults. A house colour code, a template workbook, a sheet layout the user has described: those outrank every rule here. Inside a workbook the font colour code below outranks the general house palette in
ui-design(green means a cross-sheet link here, not a gain). The rules below are for when nothing has been specified.
Decide: Which Output?
| Want | Use |
|---|---|
| A model the user will edit, extend, or plug into their own workbook | xlsx (this skill) |
| A polished document to read, print, or share, possibly with charts | html-report |
| A dataset for another program | .csv via pandas, no styling |
| One table in the conversation | markdown table |


