skills.rest
lyndonkl
lyndonklCOMMUNITY
Explore all skills by this author →
GitHub

narrative-to-numbers

Convert business narratives into defensible DCF driver inputs with auditable story links.

Updated Sep 9, 2026
Search Tags:#corporate finance#dcf valuation#intrinsic valuation#value drivers#damodaran#equity narrative#story to numbers
One-click install
npx skills add https://github.com/lyndonkl/hermesworld --skill narrative-to-numbers-lyndonkl
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: narrative-to-numbers
Source: https://github.com/lyndonkl/hermesworld/tree/main/packages/business-narrative-analyst/skills/corporate-finance/narrative-to-numbers
Command: npx skills add https://github.com/lyndonkl/hermesworld --skill narrative-to-numbers-lyndonkl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Valuations often fail because the story behind a company and the numbers in the model are disconnected: inputs appear with no justification, or narrative claims never reach a driver. This Skill enforces a two-way bridge where every story claim moves exactly one DCF driver and every model input carries one sentence of story, making assumptions testable and arguable. ## Core Features & Use Cases - Narrative grading and testing: Sorts each claim as possible, plausible, or probable, routes it to the right valuation device (option value, growth rate, or base cash flows), and screens the claim set against impossible, implausible, and improbable tests including the growth/risk/reinvestment triangle. - Driver mapping with reference classes: Maps claims to dcf-valuation-engine payload fields (revenue growth, operating margin, sales-to-capital, cost of capital, failure probability, bridge items), each anchored to industry reference menus rather than invented numbers. - Feedback loop and failure gallery: Classifies news as break, shift, or change, values counter-narratives, and documents failure patterns like the runaway story (Theranos) and the big market delusion (online advertising). - Use Case: Valuing a growth company like Tesla: write the narrative in prose, grade each claim, map five levers against auto-industry reference classes, generate drivers.json, run the DCF engine, and check marginal ROIC and terminal excess returns before trusting the per-share value. ## Quick Start Ask the agent to turn your investment story for a company into DCF driver inputs, grading each claim and producing a drivers.json payload with a story link on every row.

Frequently Asked Questions about narrative-to-numbers

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

FAQPage Schema
How do I turn a business story into DCF valuation assumptions?▼

Write the narrative in prose first, break it into discrete claims, grade each as possible, plausible, or probable, then map every claim to exactly one driver such as revenue growth, operating margin, or sales-to-capital. Each driver gets a one-sentence story link so the model stays auditable.

How do I choose a target operating margin for a DCF model?▼

Anchor the target margin on mature firms with the same business model and state which percentile of that reference class you picked, then converge to it over a stated number of years. Never anchor on a loss-making current margin or pick a number with no reference class.

What is the difference between possible, plausible, and probable claims in valuation?▼

Probable claims have evidence and enter base cash flows, plausible claims are reasoned arguments that raise the expected growth rate, and possible claims cannot be assigned a probability so they are valued as options on top of the DCF. Each claim is routed exactly once to avoid double counting.

When should a company valuation include a probability of failure?▼

Include a failure probability for young firms, money losers, and highly leveraged companies, weighting the going-concern DCF by survival probability and adding distress proceeds. Never combine a failure probability with a distress-adjusted discount rate, since that double counts the risk.

Why does my DCF value change so much when one assumption moves?▼

Large sensitivity to a single input is normal for growth companies because most value sits in the terminal year. The correct response is to show a scenario grid with likelihood labels and name the cell you believe, not to hide the sensitivity.

How do I check if my valuation narrative is internally consistent?▼

Run the impossible screens (terminal growth at or below the riskfree rate, terminal reinvestment equal to g/ROC), compute marginal ROIC against industry leaders, and check the terminal excess return. Any terminal return on capital above the terminal cost of capital claims a perpetual moat that must be named in the prose.

Related Skills

Explore More Finance & Accounting Skills→
aiunlocked1412aiunlocked1412
14

financial-analyst

Analyze financial performance and value companies using DCF valuation and financial modeling.

Community
Advanced
severin-yeseverin-ye
3

US Stock Evaluation

Evaluate US stocks via fundamental review, DCF valuation, and risk assessment.

Community
Advanced
prinzevalprinzeval

risk-analysis

Compute VaR, CVaR, and maximum drawdown across historical and Monte Carlo scenarios.

Community
Advanced
monarchjunomonarchjuno
296

aswath-damodaran

Translate company narratives into intrinsic valuation assumptions and test market expectations.

Community
Intermediate