whale-hunting
Detect institutional whale activity in options for a given underlying. Use when the user asks about unusual options activity, large block trades, whale trades, or institutional options flow for a specific symbol.
How do I install this agent skill?
npx skills add https://github.com/staskh/trading_skills --skill whale-huntingIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The whale-hunting skill is a legitimate financial tool for detecting institutional options activity. It follows security best practices by using environment variables for API keys and restricting logic to data analysis of financial tickers.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
- ZeroLeakspass
Score: 93/100 · 2 sections analyzed
What does this agent skill do?
Whale Hunting
Scans option chains for a given underlying to identify institutional-sized trades using a two-step approach:
- Crude scan (Yahoo Finance) — finds contracts with anomalous daily investment vs the rest of the chain.
- Precise drill-down (Massive API) — fetches per-second bars for each candidate and flags seconds with outlier dollar invested. Minimum whale order is $50k per transaction; in thin contracts (fewer than 10 traded seconds in the day) any second averaging more than $50k per transaction is a whale.
The scan always covers a two-session window: the latest trading session plus the one before it. During market hours that means today so far plus the full previous day. Past dates are not supported because Yahoo option chains are a live snapshot.
Instructions
Note: If
uvis not installed orpyproject.tomlis not found, replaceuv run pythonwithpythonin all commands below.
uv run python .claude/skills/whale-hunting/scripts/whale_hunting.py SYMBOL [--months N] [--sigma-z F] [--summary]
Arguments
SYMBOL— Underlying ticker (e.g.AAPL,NVDA,SPY)--months— Max months until option expiration to consider (default: 2)--sigma-z— Modified Z-Score threshold for outlier detection in both the crude scan and the per-second drill-down (default: 3.5)--summary— Also compute per-ticker summary and include it in the JSON output
Output
Returns JSON with:
underlying— The scanned symboltrading_date— Latest trading sessionwindow_dates—[previous session, latest session]; all totals cover bothsource—"massive"(per-second data) or"yahoo"(daily chain data fallback)total_whales— Total whale events foundfailed_fetches— Per-second drill-down requests (one per contract per session) that errored; each is logged to stderrtotal_call_invested— Sum of invested dollars in call whale eventstotal_put_invested— Sum of invested dollars in put whale eventscall_put_ratio— Call invested / put invested (null if no puts)put_call_ratio— Put invested / call invested (null if no calls)whale_call_wall/whale_put_wall—{strike, invested}: the strike with the most whale dollars per side (null if none)open_interest— Positioning from every fetched chain (expiries up to--months), not only whale contracts:call_oi,put_oi,put_call_oi_ratio(null if no call OI)call_wall/put_wall—{strike, open_interest}: the strike with the most open interest per side, summed across expiries (null if none)
directional_call_invested— Call invested where break-even is ≥ 1.5% above spotdirectional_put_invested— Put invested where break-even is ≥ 1.5% below spotdirectional_call_put_ratio— Directional call / directional put invested (null if no directional puts)whales— List of whale events:timestamp,ticker,type,strike,spot(underlying price at trade time),expiryclose,volume,transactions,invested,break_even,pct_from_spot(break-even distance from spot, %)- Missing values (e.g.
spotfor a trade before the first underlying bar) arenull
summary(present only when--summaryis passed) — List of per-ticker aggregates:ticker,type,strike,expiry,whale_count,total_invested,break_even
Examples
# Hunt whales for AAPL (latest and previous trading sessions)
uv run python .claude/skills/whale-hunting/scripts/whale_hunting.py AAPL
# With per-ticker summary
uv run python .claude/skills/whale-hunting/scripts/whale_hunting.py HOOD --months 3 --summary
# Looser detection threshold
uv run python .claude/skills/whale-hunting/scripts/whale_hunting.py SPY --sigma-z 2.0
Reporting
After running the script, present the results as follows.
Header line:
Whale activity for {underlying} on {window_dates[0]} – {window_dates[1]} — source: {source} Call flow: ${total_call_invested:,.0f} | Put flow: ${total_put_invested:,.0f} | C/P ratio: {call_put_ratio:.2f} Directional (≥1.5% B/E from spot) — Calls: ${directional_call_invested:,.0f} | Puts: ${directional_put_invested:,.0f} | C/P ratio: {directional_call_put_ratio:.2f} Open interest — P/C ratio: {open_interest.put_call_oi_ratio:.2f} | Call wall: {open_interest.call_wall.strike} | Put wall: {open_interest.put_wall.strike}
Print — for any null ratio or wall.
When --summary was requested, render the summary array as a table:
| Time (ET) | Ticker | Type | Strike | Expiry | # Events | Total Invested | Break Even |
|---|---|---|---|---|---|---|---|
| {timestamp} | {ticker} | {type} | {strike} | {expiry} | {whale_count} | ${total_invested:,.0f} | {break_even} |
Sort by total_invested descending. For multi-event rows use the time range of first–last event (e.g. 11:46–12:33).
Interpretation guidance:
source: "massive"— High-confidence; per-second block trade data from Massive APIsource: "yahoo"— Fallback; daily-level data from Yahoo chains (Massive API key missing, or the drill-down found no whales). Nospot/pct_from_spot, so directional totals are 0- Judge direction by
directional_call_put_ratio, notcall_put_ratio— trades with break-even within 1.5% of spot (deep ITM stock replacement, hedges) carry no directional signal - Low directional C/P ratio (< 0.5) — Bearish institutional positioning
- High directional C/P ratio (> 2.0) — Bullish institutional positioning
- OI call wall often acts as resistance and put wall as support into expiry; P/C OI > 1 leans defensive
transactions: 1— Single block trade; strongest whale signalfailed_fetches > 0— Totals may be understated; mention the count alongside the header line
Requirements
MASSIVE_API_KEYenvironment variable for per-second data. Without it, falls back to Yahoo Finance daily data (source: "yahoo"). An invalid key or a plan without 1-second aggregates is an error, not a fallback.
Timezone
All timestamps and time-based calculations must use the America/New_York timezone. All JSON output must include generated_at (NY time string) and data_delay fields.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/staskh/trading_skills/whale-hunting">View whale-hunting on skillZs</a>