Etf PremiumSAFE

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: etf-premium
description: >
  Calculate an ETF's premium or discount to NAV from Yahoo Finance data (yfinance),
  compare or screen ETFs by premium, explain why a gap exists, and decompose a sudden
  ETF move into NAV-driven vs structural components (dealer gamma exposure, blocked AP
  arbitrage, sentiment). Use this skill whenever the user asks whether an ETF trades
  above or below NAV, compares ETF premiums or discounts, screens for the biggest
  ones, asks about ETF arbitrage or premium convergence, or wants to know why an ETF
  jumped or diverged from its holdings — including gamma squeezes, dealer gamma
  exposure (GEX), and blocked creation/redemption. Especially relevant for leveraged,
  inverse, international, bond, commodity, and crypto ETFs (IBIT, BITO, HYG, KWEB).
---

# ETF Premium/Discount Analysis Skill

Calculates the premium or discount of an ETF's market price relative to its Net Asset Value (NAV) using data from Yahoo Finance via [yfinance](https://github.com/ranaroussi/yfinance).

**Why this matters:** An ETF's market price can diverge from the value of its underlying holdings (NAV). When you buy at a premium, you're overpaying relative to the assets; at a discount, you're getting a bargain. This divergence is typically small for liquid US equity ETFs but can be significant for bond ETFs, international ETFs, leveraged/inverse products, and crypto ETFs — especially during periods of market stress.

**Important**: For research and educational purposes only. Not financial advice. yfinance is not affiliated with Yahoo, Inc.

---

## Step 1: Ensure Dependencies Are Available

**Current environment status:**

```
!`python3 -c "exec('try:\n import yfinance, pandas, numpy\n print(f\'yfinance={yfinance.__version__} pandas={pandas.__version__} numpy={numpy.__version__}\')\nexcept Exception:\n print(\'DEPS_MISSING\')')"`
```

If `DEPS_MISSING`, install required packages:

```python
import subprocess, sys
subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "yfinance", "pandas", "numpy"])
```

If already installed, skip and proceed.

---

## Step 2: Route to the Correct Sub-Skill

Classify the user's request and jump to the matching section. If the user asks a general question about an ETF's premium or discount without specifying a particular analysis type, default to **Sub-Skill A** (Single ETF Snapshot).

| User Request | Route To | Examples |
|---|---|---|
| Single ETF premium/discount | **Sub-Skill A: Single ETF Snapshot** | "is SPY at a premium?", "AGG premium to NAV", "BITO premium" |
| Compare multiple ETFs | **Sub-Skill B: Multi-ETF Comparison** | "compare bond ETF discounts", "which has bigger premium IBIT or BITO", "rank these ETFs by premium" |
| Screener / find extreme premiums | **Sub-Skill C: Premium Screener** | "which ETFs have biggest discount", "find ETFs trading below NAV", "premium screener" |
| Deep analysis with context | **Sub-Skill D: Premium Deep Dive** | "why is HYG at a discount", "is ARKK premium normal", "ETF premium analysis with context" |
| Sudden premium surge / gamma squeeze | **Sub-Skill E: Premium Surge Decomposition** | "why did KWEB jump 13% today", "is this ETF rally driven by gamma", "decompose today's ETF move", "dealer GEX for SOXL", "how long until the premium converges" |

### Defaults

| Parameter | Default |
|---|---|
| Data source | yfinance `navPrice` field |
| Price field | `regularMarketPrice` (falls back to `previousClose`) |
| Screener universe | Common ETF list by category (see Sub-Skill C) |

---

## Sub-Skill A: Single ETF Snapshot

**Goal**: Show the current premium/discount for one ETF with context about what's normal, plus a peer comparison to show how it stacks up against similar ETFs.

### A1: Fetch and compute

```python
import yfinance as yf

# Peer groups by category — used to automatically compare the target ETF against its closest peers
CATEGORY_PEERS = {
    "Digital Assets": ["IBIT", "BITO", "FBTC", "ETHA", "ARKB", "GBTC"],
    "Intermediate Core Bond": ["AGG", "BND", "SCHZ"],
    "High Yield Bond": ["HYG", "JNK", "USHY"],
    "Long Government": ["TLT", "VGLT", "SPTL"],
    "Emerging Markets Bond": ["EMB", "VWOB", "PCY"],
    "Large Growth": ["QQQ", "VUG", "IWF", "SCHG"],
    "Large Blend": ["SPY", "VOO", "IVV", "VTI"],
    "Commodities Focused": ["GLD", "IAU", "SLV", "DBC"],
    "China Region": ["KWEB", "FXI", "MCHI"],
    "Trading--Leveraged Equity": ["TQQQ", "UPRO", "SOXL", "JNUG"],
    "Trading--Inverse Equity": ["SQQQ", "SPXU", "SOXS", "JDST"],
    "Derivative Income": ["JEPI", "JEPQ", "QYLD"],
    "Large Value": ["SCHD", "VYM", "DVY", "HDV"],
}

def etf_premium_snapshot(ticker_symbol):
    ticker = yf.Ticker(ticker_symbol)
    info = ticker.info

    # Verify this is an ETF
    quote_type = info.get("quoteType", "")
    if quote_type != "ETF":
        return {"error": f"{ticker_symbol} is not an ETF (quoteType={quote_type})"}

    price = info.get("regularMarketPrice") or info.get("previousClose")
    nav = info.get("navPrice")

    if not price or not nav or nav <= 0:
        return {"error": f"NAV data not available for {ticker_symbol}"}

    premium_pct = (price - nav) / nav * 100
    premium_dollar = price - nav

    # Additional context
    result = {
        "ticker": ticker_symbol,
        "name": info.get("longName") or info.get("shortName", ""),
        "market_price": round(price, 4),
        "nav": round(nav, 4),
        "premium_discount_pct": round(premium_pct, 4),
        "premium_discount_dollar": round(premium_dollar, 4),
        "status": "PREMIUM" if premium_pct > 0 else "DISCOUNT" if premium_pct < 0 else "AT NAV",
        "category": info.get("category", "N/A"),
        "fund_family": info.get("fundFamily", "N/A"),
        "total_assets": info.get("totalAssets"),
        "net_expense_ratio": info.get("netExpenseRatio"),
        "avg_volume": info.get("averageVolume"),
        "bid": info.get("bid"),
        "ask": info.get("ask"),
        "yield_pct": info.get("yie

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.