What problem does it solve? Building financial models in Excel by hand or pasting Python-computed values into cells produces workbooks that break when assumptions change and cannot be audited. This Skill generates .xlsx files programmatically where every derived cell is a live Excel formula, inputs are color-coded and sourced, and a Checks tab validates the model. ## Core Features & Use Cases - Banker-grade conventions: Enforces blue/black/green font coloring to distinguish hardcoded inputs, formulas, and cross-sheet links, plus cell comments documenting the source of every hardcoded number. - Formula-driven models with named ranges: Writes live Excel formulas instead of pasted values and uses named ranges for cross-sheet references like WACC. - Sensitivity tables and balance checks: Builds odd-sized sensitivity grids with a highlighted base-case center cell and a Checks tab that surfaces TRUE/FALSE ties. - Headless recalculation: Ships a script that recalculates formulas via LibreOffice so downstream tools reading with data_only=True get computed values. - Use Case: Ask for a DCF model for a company and receive an ./out/model.xlsx with an Inputs tab, projection formulas, a Checks tab, and a 5x5 WACC/terminal-growth sensitivity table. ## Quick Start Ask the agent to build a DCF valuation workbook for a company with revenue, growth, and WACC assumptions, saved as an .xlsx file in the out folder.
