This walks Claude through building investment-grade comparable company analysis spreadsheets with the kind of rigor you'd expect at a bank or PE shop. It enforces a strict data hierarchy (MCP sources like S&P Kensho first, never web search for financials), requires formulas instead of hardcoded values, and builds verification checkpoints so you catch errors before the full sheet is done. The philosophy is simple: structure the headers to force good thinking, let statistics emerge from clean formulas, and make sure anyone can audit your work six months later. Comes with detailed guidance on handling Office JS versus Python environments and adapting templates to context rather than blindly copying formats. Useful for M&A valuation, benchmarking portfolio companies, or prepping investment committee decks.
npx skills add https://github.com/anthropics/financial-services --skill comps-analysis