xlsx-author
anthropics/financial-services
Generate .xlsx files headless using Python and openpyxl for managed-agent sessions.
What is xlsx-author?
This skill produces Excel workbooks as file artifacts on disk instead of driving a live Excel application. Use it in headless or managed-agent environments where no Office app is open and you need to deliver a .xlsx file to the orchestration layer.
- Write Excel workbooks to ./out/<name>.xlsx using openpyxl
- Apply color conventions: blue for hardcoded inputs, black for formulas, green for cross-sheet links
- Create multiple sheets (Inputs, calculations, Checks) with named ranges and balance-check tabs
- Return relative file paths for artifact collection by the orchestration layer
- Follow financial-modeling conventions: no hardcodes in calculation cells, formulas only
How to install xlsx-author
npx skills add https://github.com/anthropics/financial-services --skill xlsx-author- Python environment with openpyxl library installed
- ./out/ directory (created automatically by the skill if needed)
- Bash or shell access to run Python scripts
How to use xlsx-author
- 1.Write a Python script using openpyxl to define your workbook structure, sheets, and formulas
- 2.Apply color formatting: Font(color='0000FF') for blue inputs, default black for formulas, Font(color='00B050') for green links
- 3.Create an Inputs sheet for all hardcoded values and a Checks sheet for balance verification
- 4.Call wb.save('./out/<filename>.xlsx') to write the file to disk
- 5.Return the relative path ./out/<filename>.xlsx in your final message for collection
Use cases
- Generate a DCF model or financial projection as a deliverable file in headless mode
- Produce an Excel workbook with input/calculation separation and balance checks for audit compliance
- Create multi-sheet financial models with formulas and named ranges without opening Excel
- Export structured financial data to .xlsx format when live Excel editing is unavailable
- Agents running in managed-agent (CMA) or headless environments
- Financial analysts and modelers who need file-based Excel output
- Teams using orchestration layers that collect file artifacts
xlsx-author FAQ
Use xlsx-author only in headless or managed-agent mode when no live Excel application is available. If mcp__office__excel_* tools are available (Cowork plugin mode), use those instead for interactive editing with review checkpoints.
Blue (0000FF) for hardcoded inputs, black (default) for formulas, and green (00B050) for links to other sheets or files. This mirrors the audit-xls conventions.
No. Create one model per file unless explicitly asked to append. Each xlsx-author call should produce a fresh workbook.
Include balance checks that verify key relationships: balance sheet balances, cash flow ties to cash position, and other critical reconciliations. Surface results as TRUE/FALSE.
Use formulas like =Inputs!C2 to reference cells from other sheets, and create named ranges for any value referenced from external decks or memos.
Full instructions (SKILL.md)
Source of truth, from anthropics/financial-services.
name: xlsx-author description: Produce a .xlsx file on disk (headless) instead of driving a live Excel workbook — for managed-agent sessions with no open Office app.
xlsx-author
Use this skill when running headless (managed-agent / CMA mode) and you need to deliver an Excel workbook as a file artifact rather than editing a live workbook via mcp__office__excel_*.
Output contract
- Write to
./out/<name>.xlsx. Create./out/if it does not exist. - Return the relative path in your final message so the orchestration layer can collect it.
How to build the workbook
Write a short Python script and run it with Bash. Use openpyxl:
from openpyxl import Workbook
from openpyxl.styles import Font, PatternFill
wb = Workbook()
ws = wb.active; ws.title = "Inputs"
ws["B2"] = "Revenue"; ws["C2"] = 1_250_000_000
ws["C2"].font = Font(color="0000FF") # blue = hardcoded input
calc = wb.create_sheet("DCF")
calc["C5"] = "=Inputs!C2*(1+Inputs!C3)" # black = formula
wb.save("./out/model.xlsx")
Conventions (mirror audit-xls)
- Blue / black / green. Blue = hardcoded input, black = formula, green = link to another sheet/file.
- No hardcodes in calc cells. Every calculation cell is a formula; every input lives on an Inputs tab.
- Named ranges for any value referenced from a deck or memo.
- Balance checks. Include a Checks tab that ties (BS balances, CF ties to cash, etc.) and surfaces TRUE/FALSE.
- One model per file. Do not append to an existing workbook unless explicitly asked.
When NOT to use
If mcp__office__excel_* tools are available (Cowork plugin mode), use those instead — they drive the user's live workbook with review checkpoints. This skill is the file-producing fallback for headless runs.
Related skills
More from anthropics/financial-services and the wider catalog.
3-statement-model
Complete and populate integrated 3-statement financial models with proper formula linkages between Income Statement, Balance Sheet, and Cash Flow Statement.
audit-xls
Audit spreadsheet formulas and financial-model integrity in seconds.
bond-futures-basis
Analyze bond futures basis, identify CTD bonds, and evaluate basis trading opportunities.
bond-relative-value
Analyze bond richness/cheapness via spread decomposition, yield curves, credit spreads, and scenario stress testing.
accessibility-review
Run a WCAG 2.1 AA accessibility audit on designs and pages to catch contrast, keyboard, and screen reader issues.
account-research
Research target companies and contacts against your ICP, with CRM deduplication and sourced signals.