DCF Valuation Modeler
Builds auditable Excel DCF models from financial data and valuation assumptions.
The skill requires staged user confirmation, source disclosure, cell-level provenance comments, and limits the workflow to producing a model rather than executing trades; however, it does not adequately define sensitive financial-data handling, MCP/Web permission boundaries, external data flows, rollback, or recovery procedures, so 10 points are deducted.
Formula rules, failure categories, a validation script, and troubleshooting guidance provide a plausible happy path; however, the Office JS/openpyxl branching is complex, the required recalc.py is not present in the supplied files, and static review cannot reproduce key paths. Dependency availability and diagnostic failure behavior remain incomplete, so 13 points are deducted.
The audience, DCF use cases, input sources, expected Excel output, and trigger description are fairly clear; however, non-fit cases, input schemas, Chinese-language interaction, and mainland-China network reachability are not specified, while core data retrieval may depend on authenticated overseas MCP/Web services, so 7 points are deducted.
The README supplies installation, license, maintenance, and CI context, while the skill contains layered workflow guidance, examples, and troubleshooting; however, the skill lacks its own version, changelog, explicit maintenance owner, complete dependency setup, and a clear delivery path for the referenced recalc.py, so 7 points are deducted.
The documentation specifies DCF calculations, WACC, scenarios, sensitivity tables, and Excel output, making the core task plausible; however, no representative model, execution result, or key-formula verification is supplied, the output still requires human review, and data-source and environment differences may require substantial rework, so 9 points are deducted.
A committed validation script plus repository-level plugin-validation and secret-scan workflows provide limited auditability; however, there are no committed tests covering this skill's key paths, execution logs, or third-party reproduction evidence, and static review cannot verify Excel outputs, so 6 points are deducted.
- This review uses only the supplied source files; no scripts, MCP calls, or Excel recalculation were executed, so it is not proof of model correctness.
- Before delivery, confirm that recalc.py actually exists and works, and manually review key formulas, terminal growth below WACC, sensitivity-table center values, and cached calculation results.
- Define financial-data permissions, transmission, storage, and redaction controls; dependent overseas data services may be unreachable or require additional authentication in mainland-China environments.
What it does & when to use it
This skill builds discounted cash flow models for equity valuation. It can source inputs from SEC filings, analyst reports, user data, MCP servers, and web research, then project revenue, margins, free cash flow, WACC, and terminal value. It supports Bear, Base, and Bull scenarios, sensitivity tables, and an enterprise-to-equity value bridge. The intended deliverable is a professional Excel model with documented inputs, but outputs remain subject to human review.
Reads user-provided financial information and can source SEC filings, analyst reports, stock price, beta, debt, and cash through MCP servers or web research; validates net debt, diluted shares, historical margins, and tax assumptions; builds scenario-based projections for revenue, operating expenses, EBIT, taxes, NOPAT, D&A, CapEx, changes in net working capital, and unlevered free cash flow; calculates cost of equity with CAPM and combines it with after-tax debt cost and capital structure weights to derive WACC; discounts projected cash flows using a mid-year convention; calculates terminal value with the perpetuity-growth or exit-multiple method; produces enterprise value, equity value, implied share price, implied return, and sensitivity tables; and, for standalone XLSX generation, uses Python/openpyxl followed by recalc.py for formula validation.
- An investment banking analyst needs a five-year DCF for a company with historical financials and market inputs.
- An equity researcher wants Bear, Base, and Bull valuation cases with revenue-growth and margin assumptions.
- A private-equity or corporate-finance team needs an enterprise-value-to-equity-value bridge using WACC, terminal value, net debt, and diluted shares.
- A financial modeler needs sensitivity analysis for WACC versus terminal growth, revenue growth versus EBIT margin, and beta versus the risk-free rate.
Pros & cons
- Covers the full DCF workflow from data validation through implied per-share value.
- Requires live Excel formulas instead of hardcoded calculated outputs and documents hardcoded inputs with source comments.
- Includes three scenarios, CAPM-based WACC, terminal-value checks, and three sensitivity analyses.
- Defines separate implementation paths for Office JS and standalone XLSX generation.
- Requires reliable financial and market data or available data connectors; the source does not guarantee access to them.
- Standalone XLSX generation depends on Python, openpyxl, LibreOffice, and recalc.py, while Office JS requires a live Excel environment.
- The workflow requires staged user confirmation rather than assuming all forecasting inputs.
- The source provides no test suite, execution sample, or verified cross-platform results.
How to install
In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select the relevant agent or plugin. In Claude Code, run claude plugin marketplace add anthropics/financial-services followed by claude plugin install pitch-agent@claude-for-financial-services. The repository contains many skills; this profile covers only plugins/agent-plugins/pitch-agent/skills/dcf-model/SKILL.md.
How to use
After installation, trigger the workflow with /dcf or ask for a request such as “Build a DCF model for XYZ with Bear/Base/Bull scenarios, WACC, terminal value, and sensitivity analysis.” Provide a company name or ticker, historical financials, and growth assumptions. The workflow calls for staged confirmation after data retrieval, revenue projections, the free-cash-flow build, WACC, terminal value, and sensitivity analysis.
Compared to similar skills
Within the same repository, lbo-model, 3-statement-model, and comps-analysis address different modeling tasks. This skill is specifically for DCF valuation and should not be presented as an LBO model, three-statement model, or comparable-company analysis tool.