What does this skill do, and when should you use it?
This skill completes existing three-statement financial model templates covering the income statement, balance sheet, and cash flow statement. It requires historical actuals and assumption drivers to be hardcoded while projections, roll-forwards, subtotals, and cross-statement links use Excel formulas. The guidance covers template mapping, forecasting, scenario analysis, margins, credit metrics, and balance, cash, and linkage checks. It is a fit for analysts working inside an existing Excel framework, but the source provides no test suite or concrete template example.
It instructs the agent to inspect workbook tabs, sections, periods, units, input cells, formulas, named ranges, and dependencies before entering data. Depending on the environment, it uses Office JS to write formulas inside Excel or Python/openpyxl to create a standalone .xlsx file, with recalc.py run before delivery in the latter workflow. It builds or preserves forecast and roll-forward relationships for revenue, expenses, profitability, balance-sheet items, cash flow, working capital, debt, taxes, and equity, and adds margin or credit analysis only when requested or required by the template. It also requires staged review of historicals, the projected income statement, the balance sheet, and the cash flow statement, with checks for balance-sheet balance, cash tie-out, net-income linkage, retained earnings, and scenario hierarchy.
- An investment-banking analyst needs to fill historicals and forecast formulas into a client-provided three-statement Excel template.
- A financial modeler inherits a partially populated income statement, balance sheet, and cash flow framework and needs to restore integrated links.
- A private-equity or corporate-finance team needs Base, Upside, and Downside scenarios and wants to verify that each scenario balances.
- A research analyst needs to populate a three-statement template from SEC filings when the template specifically requires 10-K or 10-Q data extraction.
- A model reviewer needs to test balance-sheet balance, cash consistency, net-income links, and hardcode-free projection formulas.
How do you install this skill?
- Office JS, openpyxl, and recalculation were not executed; validate formulas, recalculation, and all cross-statement checks on a copy before delivery.
- Retained-earnings treatment of SBC is inconsistent across the documents; establish the accounting convention and reconcile the formulas.
- The workflow handles financial data and may use SEC data, but does not define redaction, access control, source logging, or network-failure handling.
- No explicit rollback or versioned-delivery procedure is provided; preserve the original template and log formula changes.
- Shell / CLI
- Local filesystem
Excel or Office JS for in-Excel operationPythonopenpyxlrecalc.py for standalone .xlsx recalculation
The skill is stored at plugins/agent-plugins/model-builder/skills/3-statement-model/ and is bundled with the repository's Model Builder agent. To install the collection according to the README, run claude plugin marketplace add anthropics/financial-services, followed by claude plugin install financial-analysis@claude-for-financial-services. Alternatively, in Cowork open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select the relevant agent or vertical plugin. The source does not document a command for installing this skill alone.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/agent-plugins/model-builder/skills/3-statement-model" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
In a session where the skill is available, provide an existing three-statement Excel template and use a prompt such as: Populate this three-statement financial model template. First map the tabs and sections and ask for confirmation, then complete historicals, the income statement, balance sheet, and cash flow statement step by step. Use formulas for every projection cell. Follow the prescribed checkpoints: confirm the template map, historical block, projected income statement, balance-sheet checks, and cash-flow tie-out. Do not request an end-to-end completion that skips these reviews. For in-Excel operation, use Office JS; for a standalone .xlsx, use Python/openpyxl and run recalc.py.
What are this skill's strengths and limitations?
- Explicitly enforces formulas for projections, roll-forwards, subtotals, and cross-statement links instead of precomputed hardcodes.
- Covers template mapping, three-statement integration, scenarios, credit metrics, and extensive audit checks.
- Provides separate operating principles for Office JS and Python/openpyxl workflows.
- Uses staged confirmation and statement-by-statement validation to catch errors before final delivery.
- It is primarily a procedural modeling guide; the source includes no runnable implementation, concrete template, or test results.
- The standalone .xlsx workflow depends on Python, openpyxl, and recalc.py, which is not described in detail in the supplied material.
- Margins and credit analysis are conditional features and should be skipped unless requested or required by the template.
- The user still has to determine the appropriate data sources, accounting conventions, and line-item mapping for the supplied template.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Three-Statement Model Builder this page ✓ Anthropic · Official | 55 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| 3-Statements-Ultra | 55 · Use with care | ★ 85 | 5mo ago | — |
| XLSX Spreadsheet Skill (Open Cowork) | 69 · Recommended | ★ 2.2k | 12d ago | MIT |
| Model Checker for Financial Audits | 53 · Use with care | ★ 1.8k | 11d ago | Apache-2.0 |
| 3-Statement Financial Model Skill | 51 · Use with care | ★ 1.8k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill explicitly favors formulas, staged user confirmation, and restricts hardcoded values to historicals and assumptions, reducing accidental overwrites; it lacks sensitive-financial-data handling, permission isolation, rollback, external data-flow disclosure, and delivery recovery procedures, so points were deducted.
It documents Office JS and openpyxl paths, formula checks, common spreadsheet errors, and cross-statement validation, making the happy path fairly clear; however, recalc.py is not supplied, no reproducible tests or key-path execution evidence are present, and retained-earnings treatment of SBC is inconsistent across sections, so points were deducted.
Triggers, intended modeling tasks, Excel environments, and major workflow variants are reasonably clear; input/output contracts, Chinese-language support, mainland-China reachability, and non-fit boundaries are not sufficiently specified, and the workflow depends on Excel/Python capabilities, so points were deducted.
The documentation is layered and includes formatting, formula, and SEC-reference material; the repository supplies Apache-2.0 licensing, contribution guidance, official provenance, and CI validation context. The skill itself lacks versioning, changelog, named maintenance ownership, installation/dependency notes, and FAQs, and referenced-file availability is not established inside the skill, so points were deducted.
The stated goal is to populate and integrate three financial statements, with formulas, formatting, scenario analysis, and audit checks that cover the core task on static inspection; no representative artifact, execution result, or comparative-benefit evidence is provided, and human review remains necessary, so the static ceiling applies.
Formula references, check definitions, and repository CI provide some auditability, along with source and license context; however, CI validates plugin structure rather than this skill's modeling paths, and there is no committed test suite or third-party execution evidence, so points were deducted.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →