Company ValuationSAFE

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: company-valuation
description: >
  Estimate a public company's intrinsic value with DCF, relative (peer multiple), and
  sum-of-the-parts (SOTP) methods, then blend them into an implied share price with
  upside/downside vs the market price, a WACC and terminal-growth sensitivity grid,
  and bull/base/bear scenarios. Use this skill whenever the user asks what a company
  or ticker is worth: fair value, intrinsic value, implied share price, a price target
  from fundamentals, whether it is overvalued or undervalued, building a DCF (WACC,
  terminal value, discounted cash flow), EV/EBITDA or P/E based targets, peer
  comparison valuation, or SOTP and conglomerate discounts. Run the model rather than
  answering valuation questions from memory.
---

# Company Valuation

Triangulates intrinsic value via three methods, then blends them to an implied share price:

1. **DCF** — 5-year FCFF projection, discount at WACC, terminal value.
2. **Relative** — apply peer median P/E, EV/Revenue, EV/EBITDA.
3. **SOTP** — when 2+ distinct reporting segments exist, value each at pure-play peer multiples.

Always present a WACC × terminal-growth sensitivity table and Bull/Base/Bear scenarios.

**Disclaimer**: Research/educational output. Not financial advice.

---

## Step 1: Detection Flow

Detect data source and runtime deps. The skill supports 2 method paths — pick the richest one available.

**Environment status:**

```
!`python3 -c "exec('try:\n import yfinance, numpy, pandas\n print(\'YFIN_OK\')\nexcept Exception:\n print(\'YFIN_MISSING\')')"`
```

```
!`python3 -c "exec('try:\n import yfinance as yf\n t=yf.Ticker(\'^TNX\')\n p=t.fast_info.last_price\n print(f\'RF_10Y={p/100:.4f}\')\nexcept Exception:\n print(\'RF_FETCH_FAIL\')')"`
```

**Decision tree:**

| Condition | Method path |
|---|---|
| `YFIN_OK` | **Path A** (primary): yfinance for financials + peer multiples |
| `YFIN_MISSING` | **Path B**: pip-install yfinance, then Path A. `python3 -m pip install -q yfinance numpy pandas` |
| `RF_FETCH_FAIL` | Use default `rf = 0.045` and note stale risk-free rate in output |

If `RF_10Y=` printed, use that value as `rf` in Step 4d instead of the hardcoded 4.5%.

---

## Step 2: Choose Methods & Set Defaults

### Method applicability

| Company type | DCF | Relative | SOTP | Fallback |
|---|---|---|---|---|
| Mature cash-flow (CPG, telecom, utilities) | ✅ primary | ✅ | ❌ | — |
| High-growth SaaS / software | ✅ with care | ✅ primary | ❌ | Use EV/Revenue + Rule of 40 |
| Multi-segment conglomerate | ✅ | ✅ | ✅ primary | See `references/sotp.md` |
| Banks / insurance | ❌ | ✅ (P/B, P/TBV) | ❌ | DDM or excess return; note in output |
| Pre-revenue | ❌ | EV/Revenue only | ❌ | Flag low confidence |
| REITs | ❌ | ✅ (P/FFO, P/AFFO) | ❌ | NAV-based |
| Cyclicals (energy, semis, industrials) | ✅ on mid-cycle | ✅ | sometimes | Normalize through-cycle |

### Defaults table

Settle every parameter before pulling data — these defaults apply unless the user overrides them.

| Parameter | Default | Rationale |
|---|---|---|
| Projection horizon | 5 years | Standard explicit forecast window |
| Terminal growth `g` | 2.5% | ~ long-run US GDP |
| Risk-free rate `rf` | Live 10Y UST from Step 1, else 4.5% | Current cost of capital anchor |
| Equity risk premium `erp` | 5.5% | Damodaran mid-range |
| Beta | `info['beta']` from yfinance | Market-observed levered beta |
| Cost of debt `kd` | `interest_expense / total_debt`, else 5.5% | Effective rate; fallback to IG spread |
| Tax rate | 3-yr median effective rate, floored 15%, capped 30% | Strips out one-offs |
| Margin assumptions | 3-yr median of each ratio | Smooths cyclical noise |
| SBC treatment | Cash for software/SaaS; non-cash for industrials/CPG | Industry convention |
| Peer count | 4-6 | Balances signal vs noise |
| Peer multiple | Median (not mean) | Robust to outliers |
| Method weights (no SOTP) | DCF 50% / Relative 50% | Equal triangulation |
| Method weights (with SOTP) | DCF 40% / Relative 30% / SOTP 30% | SOTP gets weight when applicable |
| Sensitivity grid | WACC ±1% in 0.5% steps × g from 1.5-3.5% in 0.5% | 5×5 matrix |

See `references/wacc_erp_rates.md` for current risk-free rates, ERP tables, and sector WACC benchmarks.

---

## Step 3: Pull Data

```python
import yfinance as yf
import numpy as np
import pandas as pd

TICKER = "AAPL"  # replace
t = yf.Ticker(TICKER)

info       = t.info
income_a   = t.income_stmt
cashflow_a = t.cashflow
balance_a  = t.balance_sheet
income_q   = t.quarterly_income_stmt
cashflow_q = t.quarterly_cashflow

earnings_est = t.earnings_estimate
revenue_est  = t.revenue_estimate

price       = info.get("currentPrice") or info.get("regularMarketPrice")
market_cap  = info.get("marketCap")
shares_out  = info.get("sharesOutstanding")
total_debt  = info.get("totalDebt") or 0
cash        = info.get("totalCash") or 0
beta        = info.get("beta") or 1.0
sector      = info.get("sector")
industry    = info.get("industry")
```

Key financial statement rows (yfinance labels):

| Need | Row |
|---|---|
| Revenue | `Total Revenue` |
| EBIT | `Operating Income` |
| Net income | `Net Income` |
| D&A | `Depreciation And Amortization` (in cashflow) |
| CapEx | `Capital Expenditure` (negative) |
| ΔNWC | `Change In Working Capital` (cashflow) |
| SBC | `Stock Based Compensation` (cashflow) |

---

## Step 4: DCF Build

Full methodology + industry-specific tweaks in `references/dcf.md`. Quick skeleton:

```python
# 4a. Revenue growth path — fade from Y1 (consensus or hist CAGR) to terminal g
rev = income_a.loc["Total Revenue"].dropna().iloc[::-1].values  # yfinance columns are newest-first; reverse to oldest -> newest
hist_cagr = (rev[-1] / rev[0]) ** (1 / (len(rev)-1)) - 1
y1 = float(revenue_est.loc["+1y", "growth"]) if "+1y" in revenue_est.index else hist_cagr
g_terminal = 0.025
growth_path = np.linspace(y1, g_terminal + 0.01, 5)

# 4b. Margins — 3y median
ebit_margin = float((income_a.loc["Operating Income"] / income_a.loc["Total Reven

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.