Three-Statement Model Builder
Complete existing three-statement templates with linked formulas, projections, scenarios, and audit checks.
The skill requires formula-driven projections, staged user confirmation, and hardcodes only for historicals and assumptions, reducing silent overwrites and uncontrolled changes. It does not specify least-privilege access, sensitive financial-data handling, data-flow disclosure, rollback, or dependency security; the SEC guidance also points to external services, so points are deducted.
The Office JS and openpyxl paths, validation checklist, error categories, and debugging workflow are reasonably coherent. However, the supplied files contain no execution evidence or key-path test coverage, and recalc.py is not provided in the evidence; abnormal-input and dependency-failure feedback is therefore incomplete, so the score stays below the static ceiling.
The audience and triggers are clear, and the skill covers Excel execution and standalone xlsx generation as well as template, scenario, and conditional SEC-data workflows. Non-fit boundaries and input/output contracts are incomplete, Chinese-language support is unspecified, and SEC use depends on overseas services, so points are deducted.
The documentation is well layered across formatting, formulas, SEC extraction, and references; the repository supplies Apache-2.0 licensing, plugin-validation CI, and contribution guidance. The skill lacks its own version, changelog, explicit maintenance owner, and concrete installation/dependency notes, and referenced files/scripts are not fully proven available by the supplied evidence.
The skill gives detailed, directly actionable guidance for integrated three-statement modeling, scenarios, and quality checks. There is no representative completed workbook, execution proof, or directly deliverable output example, and manual template adaptation and review remain necessary; the static score is therefore capped at 7.
Pinned repository CI, formula references, check formulas, and SEC mapping provide some auditability. There is no skill-specific test suite, execution log, or independent third-party corroboration covering the key paths, so the static score is limited to 5.
- Static documentation cannot establish that formulas work correctly in real templates, circular-reference cases, abnormal inputs, or different Excel environments.
- Before delivery, verify that recalc.py, referenced files, openpyxl/Office JS runtimes, and the target workbook actually exist and are usable.
- Before handling SEC or corporate financial data, define authorization, sensitive-data protection, external transfer, and network-access controls; do not use unverified model outputs directly for investment, tax, accounting, or regulatory decisions.
- The skill requires staged user confirmation but does not define recovery or rollback when confirmation is denied, the template is corrupted, or a partial write occurs.
What it does & when to use it
This skill completes integrated Income Statement, Balance Sheet, and Cash Flow Statement templates. It reviews the workbook structure, historical and forecast periods, input cells, named ranges, and dependencies before entering data. Historical actuals and assumption drivers remain hardcoded, while projections, roll-forwards, subtotals, and cross-statement links use formulas. It also defines checks for balance-sheet balance, cash tie-out, net-income linkage, retained earnings, scenarios, and formula integrity.
Reviews workbook tabs, row and column layouts, period labels, named ranges, input cells, formulas, and dependencies; writes formulas through Office JS in Excel or through Python/openpyxl for a standalone XLSX file; populates historical data and assumptions; builds Income Statement projections, Balance Sheet roll-forwards, and Cash Flow Statement linkages; optionally adds margin analysis, credit metrics, and Base/Upside/Downside scenarios; checks balance-sheet balance, cash tie-outs, net-income links, retained earnings, formula errors, and other model-integrity conditions.
- An investment-banking analyst needs to populate historicals and forecasts in an existing three-statement template.
- A financial modeler needs to complete a partially filled workbook while preserving its layout and formulas.
- A research analyst needs projections driven by revenue, margins, working capital, CapEx, debt, or tax assumptions.
- A model reviewer needs to validate that the balance sheet balances and cash ties to the Cash Flow Statement.
Pros & cons
- Explicitly requires formulas for projection cells, roll-forwards, linkages, and subtotals.
- Covers the three core statements and key balance, cash, and net-income tie-outs.
- Includes scenario analysis, margin analysis, credit metrics, and extensive validation guidance.
- Provides both Office JS and Python/openpyxl workflows.
- It depends on an existing template, historical data, and assumptions; no sample workbook or data is included in the source.
- The standalone XLSX workflow refers to recalc.py without providing its implementation or location.
- Specific versions for Excel, Office JS, Python, or openpyxl are not documented.
- Its staged user-confirmation process may not fit an unattended end-to-end workflow.
How to install
The skill is located at plugins/agent-plugins/pitch-agent/skills/3-statement-model/ and is bundled with Pitch Agent. In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select the relevant agent; alternatively upload a zip of plugins/agent-plugins/pitch-agent/. For Claude Code, use claude plugin marketplace add anthropics/financial-services, followed by claude plugin install pitch-agent@claude-for-financial-services. The source does not document a standalone installation procedure for this individual skill.
How to use
Provide an existing three-statement Excel template and a request such as “Complete this three-statement financial model template using the supplied historical data and assumptions while preserving integrated formulas.” The skill specifies staged confirmation after template mapping, historical population, Income Statement construction, Balance Sheet construction, and Cash Flow Statement construction. The source does not specify a particular template format, data source, or location for recalc.py.