company-classification-routing

Classifies companies and routes them to the correct valuation engine and constraint set.

Updated Sep 9, 2026
One-click install
npx skills add https://github.com/lyndonkl/hermesworld --skill company-classification-routing-lyndonkl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: company-classification-routing
Source: https://github.com/lyndonkl/hermesworld/tree/main/packages/business-narrative-analyst/skills/corporate-finance/company-classification-routing
Command: npx skills add https://github.com/lyndonkl/hermesworld --skill company-classification-routing-lyndonkl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Standard valuation machinery silently produces confident but wrong numbers when pointed at banks, pre-revenue startups, distressed firms, or cyclical companies at a cycle extreme. This Skill converts what is knowable about a company into an explicit route — one primary valuation engine, overlays, an ordered pipeline, and a list of forbidden methods — before any valuation begins. ## Core Features & Use Cases - Signal extraction and gating: Computes sector type, life-cycle stage, earnings status, leverage, distress markers, and other signals from financial statements, then runs sequential gates (S2–S6) to select one of sixteen branches B1–B16. - Constraint compilation: Emits a machine-readable classification.json with hard constraints such as no-fcff-valuation for banks or require-failure-probability for distressed firms, enforced downstream by the valuation critic. - Combination rules: Resolves multi-branch companies deterministically with precedence rules, mutually exclusive pairs, and multiplicative probability composition. - Use Case: Before valuing a loss-making emerging-market bank, run the routing pipeline to produce a classification artifact that mandates an FCFE-to-regulatory-capital engine, an exposure-weighted equity risk premium, and an equity-wipeout probability — and forbids any FCFF or enterprise-multiple approach. ## Quick Start Classify this company using its financial statements and mandate, then write the classification.json route and diagnosis.md reasoning for the valuation team.

Frequently Asked Questions about company-classification-routing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the right valuation model for a difficult company?

Compute the S1 signals from the financial statements, then run the gates in order: sector gate first, then ownership and transaction, then life-cycle and earnings, then survival. The gates select exactly one engine branch such as revenue-driven for young firms or dividend discount for banks.

How do I value a bank or financial service firm?

Financial service firms route to branch B5, which forbids FCFF, WACC, and enterprise multiples because debt is raw material rather than financing. Value equity directly using a dividend discount model, FCFE to regulatory capital, or an excess-return model on book equity.

When should I normalize earnings versus use a revenue-driven valuation?

Normalization is legitimate only when losses are temporary, with evidence like a sector downturn, peers showing the same pattern, and prior normal margins. Structural, life-cycle, or leverage-driven losses route to the revenue-driven B1 branch instead, and combining both is a hard error.

Can this handle companies that fit multiple categories at once?

Yes, overlays like intangible-heavy, emerging-market, and cross-holdings compose on top of one engine branch. Precedence rules pick the single engine, mutually exclusive pairs are hard errors, and probability-weighted adjustments compose multiplicatively with each risk charged exactly once.

What are the limitations of standard DCF for distressed companies?

A DCF assumes survival to stable growth, so it overstates value when failure is possible, and raising the discount rate does not fix this. The correct repair is probability weighting: going-concern value times survival probability plus distress proceeds times failure probability.