Skip to content

JPeetz/agent-skills / skills/finance-ops/financial-modeling

financial-modeling

Verified

Use this skill when building financial models — three-statement models, bottoms-up P&L, scenario planning, forecasting, or valuation. Activates on "financial model", "three-statement model", "P&L", "balance sheet", "cash flow forecast", "scenario planning", "pro forma", "unit economics", "bottoms-up", "top-down forecast", "sensitivity analysis", "SaaS model", "fundraising model", "cap table", "DCF", or "runway model". Covers driver-based revenue builds, COGS and OPEX structure, P&L → balance sheet → cash flow links, best/base/worst scenarios, tornado sensitivity, DCF and terminal value, dilution and liquidation waterfalls, and the checks that catch a model that does not foot.

FileKindSize
CHANGELOG.mdmarkdown2.8 KB
LICENSEother1.0 KB
SKILL.mdskill md28.5 KB
evals/evals.jsonconfig11.5 KB
references/model-framework.mdmarkdown14.2 KB
scripts/financial_model_builder.pyscript53.7 KB

CHANGELOG.md

View on GitHub
# Changelog

All notable changes to this skill are documented here. Format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), versioning follows [SemVer](https://semver.org/).

## [1.0.0] — 2026-11-16

### Added
- `SKILL.md` — the full modelling stack: driver-tree discipline, bottoms-up P&L construction by channel (with rep ramp and seasonality), COGS decomposition, headcount-led OPEX with burden and hire lag, the P&L → balance sheet → cash flow link table and build order, scenario design as a single driver switch, tornado and breakeven sensitivity, unit economics on gross-margin dollars, SaaS stock-and-flow modelling with cohort decay, DCF with terminal-value cross-checks, fundraising mechanics (pre/post-money, option-pool shuffle, stacked SAFEs, liquidation waterfall), the twelve metrics every model must emit, a data-quality checklist, a seventeen-row anti-pattern table, an investor-presentation order, and a verification checklist.
- `scripts/financial_model_builder.py` — stdlib-only three-statement engine taking a JSON spec and emitting monthly P&L, cash flow and balance sheet for 1–60 months plus annual roll-up. Driver-based MRR roll-forward with seasonality, headcount plan by department with per-month additions and COGS/OPEX bucketing, per-vintage straight-line depreciation, working capital from DSO/DPO/deferred-revenue drivers, debt schedule, tax with NOL carry-forward, indirect cash flow, derived balance sheet with a per-month balance residual, twelve operating metrics, severity-tagged flags, and a `--self-test` that verifies twenty invariants including six negative input validations. Exits non-zero on bad input and warns (exit 3) when the model does not foot.
- `references/model-framework.md` — the four model types (three-statement, bottoms-up, top-down, cohort), fit by company stage from pre-seed to pre-IPO, the fourteen-step three-statement build sequence, the six-step revenue build, the distinction between scenario/sensitivity/stress test, ten model-architecture rules, and a structural-error table by model type.
- `evals/evals.json` — four positive and two negative evaluation cases covering the three-statement build, survivable scenario planning, fundraising dilution with pool shuffle and SAFEs, sensitivity ranking, refusal to back-solve to a target, and the audit/tax/investment-advice boundary.
- `LICENSE` (MIT), `CHANGELOG.md`.

### Notes
- Complements rather than duplicates `saas-metrics-coach`: that skill computes and interprets a metrics pack from historical data, this one builds the forward-looking model that produces those metrics.
- The balance-residual check, cash-to-statement tie, ARR-to-MRR tie and retained-earnings tie are hard gates in the script, not advisory warnings, because a model that does not foot has no evidentiary value.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall