AlphaGBM Alerts

Set intelligent alerts based on price, IV rank, unusual activity, earnings timing, and VRP signals -- each alert fires with full context so you can act immediately.

What This Skill Does

How to Use

Input: An alert configuration command specifying ticker, condition, and threshold.

Output:

• Alert configuration confirmation with summary of what will be monitored
• When triggered: alert notification with full context (what triggered, current values, suggested action)
• Alert management: list active alerts, edit conditions, delete alerts

Example Queries:

• alert me when AAPL IV rank above 80 — IV rank threshold alert
• notify if NVDA drops below 850 — Price level alert
• earnings alert for TSLA — Alert 7 days before TSLA earnings
• VRP alert AAPL — Notify when AAPL VRP signal changes
• set price alert SPY 550 — Simple price target alert
• my alerts — List all active alerts
• delete alert 3 — Remove a specific alert

Mock Data

Mock data files are located in mock-data/alert/ and include:

• active-alerts.json — Sample list of configured alerts
• triggered-alerts.json — Recently triggered alerts with context
• alert-config-response.json — Example alert creation confirmation

API Endpoint

GET    /api/user/alerts
POST   /api/user/alerts
PUT    /api/user/alerts/{alert_id}
DELETE /api/user/alerts/{alert_id}
GET    /api/user/alerts/triggered

POST body:

{
  "symbol": "AAPL",
  "type": "iv_rank_above",
  "threshold": 80,
  "recurring": true
}

Response fields: alert_id, status, condition_summary, triggered_alerts[], context

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Compare

Side-by-side comparison of 2-5 stocks or options across every AlphaGBM dimension, so you can pick the best opportunity.

What This Skill Does

How to Use

Input: 2-5 ticker symbols with a comparison query.

Output:

• Comparison table with all dimensions side by side
• Winner highlighted per category (green badge)
• Overall recommendation with composite score
• Key differentiators: what makes the winner stand out
• Trade idea: if you had to pick one, which and why

Example Queries:

• compare AAPL vs MSFT — Head-to-head across all dimensions
• NVDA or AMD — Which semiconductor name is the better trade?
• which is cheaper TSLA or META options — Options cost comparison
• tech stock comparison AAPL MSFT GOOGL AMZN META — Full sector comparison
• compare options AAPL vs MSFT 30d ATM — Specific options contract comparison

Mock Data

Mock data files are located in mock-data/compare/ and include:

• aapl-vs-msft.json — Full comparison output for AAPL vs MSFT
• tech-five-way.json — Five-way comparison of mega-cap tech
• options-cost-compare.json — Options-specific metrics comparison

API Endpoint

GET /api/analytics/compare

Query parameters:

• symbols (string, required) — Comma-separated tickers (2-5), e.g., "AAPL,MSFT,GOOGL"
• dimensions (string, default "all") — Comma-separated: "pillars", "options", "technicals", "valuations"
• options_expiry (string) — Target expiry for options comparison (e.g., "30d", "60d")

Response fields: tickers[], comparison_table, category_winners, overall_ranking[], recommendation

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Earnings IV Panel

Everything you need for earnings week — historical IV crush + forward-looking implied move + IV Rank strategy recommendation + a priced Iron Condor centered on the implied move — in a single API call.

What This Skill Does

How to Use

Input: A ticker with upcoming or past earnings.

Output:

• Days to next earnings (if scheduled)
• Current stock price + ATM IV + IV Rank
• Implied Move ±X% and ±$Y — most quoted number during earnings season
• Recommendation tag (🔥 short IV / wait / directional) with zh/en copy
• Iron Condor pricing — 4 strikes + credit + max profit + max loss + breakeven bounds (Pro tier)
• Last 8 quarters: pre-earnings IV / post-earnings IV / crush % / actual move / straddle PnL
• Avg crush % and straddle win rate

Example Queries:

• earnings crush AAPL — Full crush history + next earnings IM
• implied move NVDA — What the options are pricing for next earnings
• iron condor for META — Priced-ready short-premium setup
• IV rank MSFT earnings — Strategy tag + recommendation
• should I short premium before TSLA — Recommendation + IC quote
• straddle pnl AMZN last 8 quarters — Historical short-premium win rate

Mock Data

Mock data files are in mock-data/earnings-crush/:

• aapl-crush-history.json — 8 quarters of AAPL crush + implied move + IC
• nvda-crush-history.json — Same for NVDA
• crush-summary.json — Aggregated crush statistics across tickers

API Endpoint

GET /api/options/earnings-crush/{symbol}

Query parameters:

• quarters (int, default 8) — Number of past earnings to analyze
• include_straddle_pnl (bool, default true) — Include straddle P&L simulation
• include_iron_condor (bool, default true) — Include Iron Condor quote (Pro tier in UI)

Response fields (headline numbers):

• next_earnings, days_to_earnings, current_atm_iv, current_stock_price
• implied_move_pct — e.g. 5.1 means market prices ±5.1% move
• iv_rank_pct — 0-100 percentile; feeds recommendation.level
• recommendation — {level: 'high'|'mid'|'low'|'unknown', iv_rank_pct, recommendation_zh, recommendation_en}
• iron_condor — {short_call, long_call, short_put, long_put, credit, max_profit, max_loss, breakeven_up, breakeven_down, wing_width_pct}
• crush_history[], avg_crush_pct, avg_actual_move_pct, straddle_win_rate
• quarters_analyzed, timestamp

Pricing: 1 option-analysis credit per call; cache hits (same symbol/params within 5 min) are free.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Market Sentiment Dashboard

Aggregates market-wide sentiment indicators into a single dashboard, classifying the current regime to guide your trading stance.

What This Skill Does

How to Use

Input: A market sentiment query (no ticker required, or specify VIX/SPX for focused analysis).

Output:

• Sentiment dashboard with all indicators and their current readings
• Historical context: where each indicator sits relative to the past 1 year
• Current regime classification (risk-on / risk-off / neutral) with confidence %
• Sector rotation map: early cycle, mid cycle, late cycle, or recession positioning
• Actionable interpretation: what the current sentiment means for options trading

Example Queries:

• market sentiment — Full dashboard with all indicators
• is the market fearful — Quick fear/greed assessment
• VIX analysis — Deep dive on VIX level, term structure, and percentile
• put call ratio — Equity and index put/call with historical context
• market breadth — Advance/decline, new highs/lows, participation analysis
• sector rotation — Which sectors are leading and what cycle stage we are in

Mock Data

Mock data files are located in mock-data/market-sentiment/ and include:

• sentiment-dashboard.json — Full dashboard snapshot with all indicators
• vix-history.json — VIX time series with percentile ranks
• sector-rotation.json — Sector performance and cycle classification

API Endpoint

GET /api/analytics/market-sentiment

Query parameters:

• indicators (string, default "all") — Comma-separated list: "vix", "pcr", "fear_greed", "breadth", "rotation"
• lookback_days (int, default 252) — Historical period for percentile calculations

Response fields: vix, put_call_ratio, fear_greed_index, breadth, sector_rotation, regime, regime_confidence

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Options Score

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Scores every option contract in a chain using a multi-factor model across 4 strategy types, so you instantly know which contracts have the best risk/reward profile.

Strategy Scoring Models

Sell Put Weights

Sell Call Weights

Buy Call Weights

Buy Put Weights

Score Scale

• 80-100: Exceptional — top-tier opportunity
• 60-79: Strong — good trade candidate
• 40-59: Average — proceed with caution
• 0-39: Poor — avoid unless hedging

Risk-Return Profiles

API Endpoints

Get Option Expirations

GET /api/options/expirations/<SYMBOL>

Option Chain Analysis -- Synchronous

POST /api/options/chain-sync
Content-Type: application/json

{"symbol": "AAPL", "expiry_date": "2026-04-17"}

Add ?compact=true for condensed response.

Response includes for each of 4 strategies (Sell Put, Sell Call, Buy Call, Buy Put):

• Top 10 recommendations sorted by score (0-100)
• Score breakdown: premium_yield, support/resistance_strength, safety_margin, trend_alignment, probability_profit, liquidity, time_decay
• ATR safety info (safety_ratio, atr_multiples, is_safe)
• Risk-return profile: style, risk_level, win_probability
• Trend analysis: direction, strength, alignment score

Option Chain Analysis -- Async

POST /api/options/chain-async
Content-Type: application/json

{"symbol": "TSLA", "expiry_date": "2026-04-17"}

Returns {"task_id": "uuid"}. Poll with: GET /api/tasks/<task_id>.

Enhanced Single-Option Analysis -- Sync

POST /api/options/enhanced-sync
Content-Type: application/json

{"symbol": "AAPL", "option_identifier": "AAPL260417C00190000"}

Enhanced Single-Option Analysis -- Async

POST /api/options/enhanced-async
Content-Type: application/json

{"symbol": "AAPL", "option_identifier": "AAPL260417C00190000"}

Reverse Score

Score a specific contract from known parameters:

POST /api/options/reverse-score
Content-Type: application/json

{"symbol": "AAPL", "option_type": "CALL", "strike": 190, "expiry_date": "2026-02-16", "option_price": 2.50, "implied_volatility": 28}

Batch Chain Analysis

POST /api/options/chain/batch
Content-Type: application/json

{"symbols": ["AAPL", "NVDA"], "expiries": ["2026-04-17", "2026-05-15"]}

Max 3 symbols x 2 expiries per request.

IV Snapshot (instant, no quota cost)

GET /api/options/snapshot/<SYMBOL>

Returns: ATM IV, IV Rank, HV 30d, VRP, VRP level.

Daily Recommendations (no auth required)

GET /api/options/recommendations?count=5

Typical Workflow

1. Get expirations: GET /api/options/expirations/AAPL
2. Quick IV check: GET /api/options/snapshot/AAPL (free, no quota)
3. Run chain analysis: POST /api/options/chain-sync with symbol + expiry
4. Drill into a specific contract: POST /api/options/enhanced-sync with option_identifier
5. Compare across tickers: POST /api/options/chain/batch for multi-symbol analysis

Quota

• Free: 1 options analysis/day
• Plus: 1,000/month
• Pro: 5,000/month
• Snapshot and recommendations endpoints cost nothing.

Output Formatting Tips

• Scores are 0-100; present top picks in a table sorted by score descending.
• Always show the score breakdown factors so users understand why a contract scored well.
• Highlight ATR safety info (is_safe flag) prominently for sell strategies.
• Include the risk-return style label (steady_income, balanced, etc.) for quick context.

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Uses realistic option chain snapshots from mock-data/.

Related Skills

• alphagbm-stock-analysis -- Analyze the underlying stock first
• alphagbm-options-strategy -- Build multi-leg strategies with top-scored contracts
• alphagbm-greeks -- Deep-dive into Greeks for a specific contract
• alphagbm-vol-surface -- See if IV is cheap or expensive across strikes

Powered by AlphaGBM -- Real-data options & research intelligence. 10K+ users.

AlphaGBM Polymarket Integration

Bridges prediction markets and options markets -- when Polymarket says 70% chance of a rate cut but options imply 55%, that is a potential mispricing you can trade.

What This Skill Does

How to Use

Input: An event type or query about prediction market vs options pricing.

Output:

• Event probability comparison table: Polymarket probability vs options-implied probability
• Probability spread and direction (which market is more bullish/bearish on the event)
• Mispricing signals ranked by confidence and spread size
• Suggested options trades to exploit the mispricing
• Historical accuracy comparison for similar past events

Example Queries:

• polymarket signals — Scan for the largest probability mismatches right now
• prediction market vs options rate cut — Compare Fed rate cut odds across markets
• event probability election — Election outcome probabilities vs options positioning
• rate cut odds — What prediction markets and options each imply about the next Fed meeting
• polymarket arbitrage — Actionable mispricing opportunities

Mock Data

Mock data files are located in mock-data/polymarket/ and include:

• rate-cut-comparison.json — Fed rate cut probabilities: Polymarket vs options-implied
• event-scan.json — Top mispricing signals across active prediction markets
• historical-accuracy.json — Past event forecasting accuracy by market type

API Endpoint

GET /api/analytics/polymarket/signals
GET /api/analytics/polymarket/event/{event_id}

Query parameters:

• event_type (string) — Filter: "fed", "election", "earnings", "macro", "all"
• min_spread (float, default 0.10) — Minimum probability spread to surface (10%)
• include_trades (bool, default true) — Include suggested options trades

Response fields: events[], polymarket_prob, options_implied_prob, spread, confidence, suggested_trades[], historical_accuracy

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Stock Analysis

Analyze stocks via the AlphaGBM API — a G = B + M (Gain = Basics + Momentum) model combining fundamental analysis, market sentiment, EV expectation, ATR stop-loss, sector rotation, and AI reports.

When to use

• User asks to analyze a stock ticker (US / HK / A-share)
• User asks for a stock quote, target price, risk score, or EV recommendation
• User mentions AlphaGBM or wants a comprehensive stock analysis

Prerequisites

• API Key: stored in env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.
• If the user has neither, tell them to register at https://alphagbm.com and create a key at /api-keys.

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. Quick Quote (instant, no quota cost)

GET /api/stock/quick-quote/<TICKER>

Returns: price, change%, PE, forward PE, 52-week range, sector, market cap.

Example:

curl -H "Authorization: Bearer $ALPHAGBM_API_KEY" \
  https://alphagbm.zeabur.app/api/stock/quick-quote/AAPL

2. Full Stock Analysis — Synchronous (blocks 10-30s)

POST /api/stock/analyze-sync
Content-Type: application/json

{"ticker": "AAPL", "style": "balanced"}

Add ?compact=true for a condensed agent-friendly response (~500 tokens).

Response contains:

• data — price, PE, PEG, growth, margin, target_price, stop_loss_price, market_sentiment (0-10), ev_model, sector_analysis, capital_analysis
• risk — score (0-10), level, suggested_position%, risk flags
• report — AI-generated narrative report (markdown, ~2000 chars)

3. Full Stock Analysis — Async (for web frontend)

POST /api/stock/analyze-async
Content-Type: application/json

{"ticker": "TSLA", "style": "growth"}

Returns {"task_id": "uuid"}. Poll task:

GET /api/tasks/<task_id>

4. Stock Search (no auth required)

GET /api/stock/search?q=AAPL&limit=8

Fuzzy search — supports US (AAPL), HK (700, 0700.HK), A-share (600519).

5. Analysis History

GET /api/stock/history?page=1&per_page=10&ticker=AAPL

6. Stock Summary (for options page linkage)

GET /api/stock/summary/<TICKER>

Returns condensed analysis. First-time analysis per ticker is free.

Analysis Model Summary

G = B + M

Risk Score (0-10, additive)

Risk 0-2 → Max position 20% · Risk 8-10 → Don't buy.

EV Expectation Model

EV = (upside_prob x upside_range) + (downside_prob x downside_range)
Weighted = 50% x 1-week + 30% x 1-month + 20% x 3-month

Target Price — 5 methods, industry-weighted

PE valuation · PEG valuation · Growth discount · DCF · Technical analysis. Risk adjustment: high risk → -15%, medium risk → -8%.

ATR Stop-Loss

stop = price - ATR(14) x multiplier(1.5-4.0)

Multiplier adjusts for Beta and VIX. Hard floor: -15%.

Typical Workflow

1. Quick check → GET /api/stock/quick-quote/NVDA
2. If interesting → POST /api/stock/analyze-sync {"ticker":"NVDA","style":"growth"}
3. Present: recommendation, target price, risk score, EV, AI report

Quota

• Free users: 2 stock analyses/day
• Plus: 1000/month · Pro: 5000/month
• Quick quote costs nothing

Output Formatting Tips

When presenting results to the user, highlight:

1. Recommendation (STRONG_BUY / BUY / HOLD / AVOID / STRONG_AVOID) + confidence
2. Target price vs current price → upside %
3. Risk score + level + top risk flags
4. Stop-loss price + method
5. EV score + weighted EV%
6. Key excerpt from AI report (first 2-3 paragraphs)

Mock Data

When no API key is configured, this skill uses built-in market data snapshots from mock-data/. Supported demo tickers: AAPL, NVDA, SPY, TSLA, META.

Related Skills

• alphagbm-options-score — After stock analysis, evaluate options opportunities
• alphagbm-compare — Compare multiple stocks side-by-side
• alphagbm-market-sentiment — Broader market context for the analysis

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Unusual Options Activity

Detects unusual options activity and classifies smart money signals to help you follow institutional positioning.

What This Skill Does

How to Use

Input: A ticker symbol or market-wide scan request.

Output:

• Unusual activity list: timestamp, strike, expiry, type (call/put), volume, OI, premium, trade classification
• Sentiment classification per trade (bullish sweep, bearish block, hedging, earnings positioning)
• Net premium flow (calls vs puts in dollar terms)
• Historical accuracy: how often similar signals preceded the expected move
• Aggregated smart money score

Example Queries:

• unusual options activity — Market-wide scan of today's most unusual trades
• smart money AAPL — Institutional flow signals for Apple
• large trades NVDA — Block and sweep orders for NVIDIA
• who's buying TSLA puts — Bearish flow analysis for Tesla
• options flow SPY — Net premium flow for S&P 500 ETF

Mock Data

Mock data files are located in mock-data/unusual-activity/ and include:

• aapl-unusual-trades.json — Recent unusual trades for AAPL
• market-wide-scan.json — Top 20 unusual activity signals across all tickers
• flow-summary.json — Aggregated premium flow by sector

API Endpoint

GET /api/options/unusual-activity/{symbol}
GET /api/options/unusual-activity/scan

Query parameters:

• min_premium (int, default 100000) — Minimum trade premium in dollars
• min_vol_oi_ratio (float, default 3.0) — Minimum volume-to-OI ratio
• trade_type (string) — Filter: "sweep", "block", "all"
• sentiment (string) — Filter: "bullish", "bearish", "all"

Response fields: trades[], net_premium_flow, sentiment_summary, smart_money_score, historical_accuracy

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Watchlist

Monitor your favorite tickers for meaningful changes -- price moves, IV shifts, unusual activity, upcoming earnings, and score changes -- all in one dashboard.

What This Skill Does

How to Use

Input: Watchlist management command or query.

Output:

• Watchlist dashboard: each ticker with current price, daily change, IV rank, next earnings date
• Alert flags: what changed since last check (price breakout, IV spike, unusual flow, etc.)
• Daily summary: priority-ranked notifications across all watchlist tickers
• Quick actions: suggested trades based on watchlist alerts

Example Queries:

• add AAPL to watchlist — Add a ticker to your custom watchlist
• my watchlist — View your full watchlist dashboard
• watch NVDA TSLA META — Add multiple tickers at once
• watchlist alerts — Show only tickers with active alerts
• remove SPY from watchlist — Remove a ticker
• hot options — View the curated high-activity options list

Mock Data

Mock data files are located in mock-data/watchlist/ and include:

• user-watchlist.json — Sample user watchlist with 10 tickers
• watchlist-alerts.json — Triggered alerts for watchlist tickers
• hot-options.json — Curated hot options list

API Endpoint

GET    /api/user/watchlist
POST   /api/user/watchlist
DELETE /api/user/watchlist/{symbol}
GET    /api/user/watchlist/alerts
GET    /api/analytics/hot-options

POST body: { "symbol": "AAPL" } or { "symbols": ["NVDA", "TSLA", "META"] }

Response fields: watchlist[], alerts[], daily_summary, hot_options[]

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM BPS Backtest

Backtests the Bull Put Spread (short put + long put at lower strike) as a mechanical strategy over 2018–present on any ticker, with two passes per call:

1. With Signal — only enters when the per-ticker FearScore is ≥ your threshold
2. No Signal (Control) — enters unconditionally every Monday

The side-by-side comparison shows whether the signal is doing work, or whether you're paying 1 credit for noise.

Parameters

All optional except ticker:

What's Returned

Per pass (with_signal and no_signal):

• total_trades, win_rate_pct, annual_return_pct, sharpe, max_drawdown_pct, roc_pct, avg_holding_days, avg_pnl_per_trade, total_pnl, final_capital
• exit_reasons — count by take_profit / stop_loss / expiry_otm / expiry_itm / close_early
• trades[] — full ledger (entry/exit date, strikes, credit, pnl, reason)
• equity_curve[] — per-day cumulative capital
• pnl_histogram — bucket counts for the P&L distribution

Plus:

• summary — one-paragraph zh/en takeaway comparing signal vs control, with ⚠️ flags when drawdown or win rate look problematic

Methodology Notes

• IV is proxied by 20-day historical volatility (HV20) for BS pricing. Historical option-chain IV is unaffordable to source at scale; HV20 is a reasonable proxy but will under-estimate IV around events. Live results typically outperform backtest because of this.
• FearScore is reconstructed from the same 6 indicators the live version uses, but computed from cheap historical price + volume data only.
• Entries filtered by max_positions (3) and min_entry_spacing_days (3) and a risk_per_trade cap (0.5% of capital).

How to Use

Example Queries:

• backtest BPS on QQQ — Default params, signal vs control comparison
• does FearScore work on SPY — Same call, reads the comparison summary
• backtest bull put spread IWM DTE 21 delta 0.30 — Custom params
• what DTE works best for BPS on QQQ — Run a few with different DTEs, compare
• bps fear threshold 70 vs 60 on NVDA — Run two calls with different thresholds

Mock Data

Mock data in mock-data/bps-backtest/ — examples for QQQ with signal ON and OFF.

API Endpoint

POST /api/options/bps-backtest
Content-Type: application/json

Request body:

{
  "ticker": "QQQ",
  "dte_target": 14,
  "short_delta": 0.25,
  "spread_width": 5.0,
  "take_profit_pct": 0.50,
  "fear_threshold": 60,
  "start_date": "2018-01-01",
  "end_date": "2026-04-20",
  "include_control": true
}

Response:

{
  "success": true,
  "ticker": "QQQ",
  "period": {"start": "2018-01-01", "end": "2026-04-20"},
  "with_signal": {
    "total_trades": 28, "win_rate_pct": 100, "annual_return_pct": 10.8,
    "sharpe": 16.3, "max_drawdown_pct": 0.0, "trades": [...], "equity_curve": [...],
    "pnl_histogram": {...}, "exit_reasons": {"take_profit": 20, "expiry_otm": 8}
  },
  "no_signal": {
    "total_trades": 185, "win_rate_pct": 82, "annual_return_pct": 3.5,
    "sharpe": 2.1, "max_drawdown_pct": -8.2, ...
  },
  "summary": {
    "zh": "QQQ · 2018-2026 · 使用 FearScore ≥ 60 触发 BPS 入场，共交易 28 笔，年化 +10.8%，胜率 100%，最大回撤 0.0%。 同参数无信号对照组年化 +3.5%、胜率 82%；信号版本高出无信号组 7.3 个百分点。",
    "en": "QQQ · 2018-2026 · BPS entry on FearScore ≥ 60 over 28 trades: annualized +10.8%, win rate 100%, max drawdown 0.0%. The no-signal control under the same params: annualized +3.5%, win rate 82%. Signal version outperforms by 7.3 pp."
  }
}

Pricing: 1 option-analysis credit per call; 30-min cache per parameter hash (cache hits free). Expect ~5-10s compute for a fresh hash.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Buffett Analysis

The 4 lenses Buffett himself says he applies, computed from yfinance fundamentals and returned as a single-number verdict plus reasoning for each lens.

The 4 Lenses

1. Business (20% weight) — circle of competence. Simple sectors (consumer staples, utilities, industrials) score high. Complex sectors (tech, healthcare, financials) score lower unless mega-cap like AAPL.
2. Moat (30% weight) — durable advantage. Gross margin > 40%, ROE > 20%, profit margin > 15%, and market cap > $100B each contribute to the moat score.
3. Management (15% weight) — capital allocation proxy via dividend continuity
   • payout ratio (15-60% is ideal balance) + 5yr avg div yield.
4. Valuation (35% weight) — fair price check. PE < 15 → +20, PEG < 1 → +15, FCF yield > 10Y treasury + 2pp → +20. PE > 40 or PEG > 2.5 → deductions.

Overall Verdict

• ≥ 75 → HOLDABLE (color green) — meets Buffett standards, long-term hold
• 55-74 → WATCHABLE (color amber) — wait for better price or clearer evidence
• < 55 → AVOID (color red) — fails Buffett's standards

Why This Is a Separate Skill

The generic alphagbm-stock-analysis runs a G=B+M style/momentum score. Buffett's framework is different — it weights moat + valuation much more heavily than momentum, and penalizes complex businesses regardless of growth. This skill codifies Buffett's rules, not AlphaGBM's house rules.

How to Use

Input:

• ticker (required) — US stock symbol

Output:

• scorecard.business: {score, sector, industry, verdict_zh, verdict_en}
• scorecard.moat: {score, gross_margin, roe, profit_margin, market_cap_b, reasons_zh, reasons_en}
• scorecard.management: {score, dividend_rate, payout_ratio, reasons_zh, reasons_en}
• scorecard.valuation: {score, pe, forward_pe, peg, pb, fcf_yield_pct, ten_year_treasury, reasons_zh, reasons_en}
• scorecard.overall: {score, verdict, verdict_zh, verdict_en, color}

Example Queries

• Buffett analysis on KO → likely HOLDABLE (simple business, strong moat, 30+ year hold by Buffett himself)
• would Buffett buy NVDA → likely WATCHABLE or AVOID (complex sector, high valuation)
• Buffett scorecard JNJ → likely HOLDABLE (consumer defensive, strong margins, reasonable PE)
• score AAPL with Buffett lens → reference Berkshire's own holding for context
• apply Buffett's checklist to WMT → retail-native test case

Mock Data

Mock data in mock-data/buffett-analysis/ — sample for KO (HOLDABLE).

API Endpoint

POST /api/masters/buffett-analyze
Content-Type: application/json

Request body:

{"ticker": "KO"}

Response shape:

{
  "success": true,
  "ticker": "KO",
  "current_price": 63.4,
  "scorecard": {
    "business": {
      "score": 85,
      "sector": "Consumer Defensive",
      "industry": "Beverages - Non-Alcoholic",
      "verdict_zh": "业务相对简单,在巴菲特能力圈范围内",
      "verdict_en": "Relatively simple business within Buffett's circle"
    },
    "moat": {
      "score": 100,
      "gross_margin": 60.3,
      "roe": 41.8,
      "profit_margin": 22.4,
      "market_cap_b": 273.4,
      "reasons_zh": ["毛利率 60.3% > 40%,显示定价权", "ROE 41.8% > 20%,资本效率强", "市值 $273B > $100B,规模壁垒", "净利率 22.4% > 15%,强定价权"],
      "reasons_en": ["Gross margin 60.3% > 40% shows pricing power", "ROE 41.8% > 20% — strong capital efficiency", "Market cap $273B > $100B — scale moat", "Net margin 22.4% > 15% — strong pricing power"]
    },
    "management": {
      "score": 80,
      "dividend_rate": 1.94,
      "payout_ratio": 77.0,
      "reasons_zh": ["派息 $1.94 — 体现向股东返现意愿", "5 年平均股息率 3.1%"],
      "reasons_en": ["Dividend $1.94 — willingness to return cash", "5-yr avg div yield 3.1%"]
    },
    "valuation": {
      "score": 45,
      "pe": 24.8,
      "forward_pe": 22.1,
      "peg": 3.2,
      "pb": 10.5,
      "fcf_yield_pct": 3.5,
      "ten_year_treasury": 4.3,
      "reasons_zh": ["FCF 收益率 3.5% < 10Y 美债 4.3%,不如债券"],
      "reasons_en": ["FCF yield 3.5% < 10Y 4.3% — bonds beat it"]
    },
    "overall": {
      "score": 78.3,
      "verdict": "HOLDABLE",
      "verdict_zh": "符合巴菲特标准 — 值得长期持有",
      "verdict_en": "Meets Buffett standards — worth a long-term hold",
      "color": "green"
    }
  },
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: 1 stock-analysis credit per call; 30-min cache per ticker (cache hits free).

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Company Profile

Build and manage company research profiles in a user's private knowledge base. Each profile captures fundamentals (PE/PB), 8-year valuation bands, financial red flags, and recent events — auto-refreshed on a schedule.

When to use

• User wants to track a company in their personal research workspace
• User asks to list / view / delete saved companies
• User asks for PE or PB historical band of a ticker
• User asks to refresh a stale profile
• User mentions "知识库" / "投研档案" / "research brain" / "knowledge base"

Prerequisites

• API Key: stored in env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.
• If the user has no key, direct them to register at https://alphagbm.com and create one at /api-keys.
• Tier requirement: Free tier = 1 profile, Plus = 10, Pro = 50. Create endpoint returns 403 with upgrade_required: true when the cap is hit.

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. List profiles

GET /api/research/profiles?page=1&per_page=20

Response:

{
  "success": true,
  "profiles": [{ "ticker": "AAPL", "company_name": "...", "current_price": 261.0, ... }],
  "total": 3,
  "page": 1,
  "per_page": 20
}

2. Get profile detail (includes thesis if one exists)

GET /api/research/profiles/<TICKER>

Returns the full profile + an embedded thesis field (null if no thesis yet). See Response schema below for field list. Returns 404 if the ticker isn't in the user's knowledge base.

3. Create profile

POST /api/research/profiles
Content-Type: application/json

{"ticker": "AAPL"}

Behavior: Pulls fundamentals via the data provider, computes red flags and event radar, persists the profile. If a profile already exists for this user+ticker, it's updated in place (idempotent).

Tier limit response (403):

{
  "success": false,
  "error": "Profile limit reached. Upgrade to Plus for 10 profiles.",
  "current": 1,
  "max": 1,
  "upgrade_required": true
}

4. Delete (archive) profile

DELETE /api/research/profiles/<TICKER>

Soft-deletes by flipping status to archived. Returns 404 if not found.

5. Refresh profile data

POST /api/research/profiles/<TICKER>/refresh

Pulls fresh market data, recomputes red flags and events. Use when the user says "refresh my profile" or the last_updated_at is stale (> 7d old).

6. PE/PB Band data (cached 24h)

GET /api/research/profiles/<TICKER>/band

Returns 8-year PE/PB history for building the band chart. Can be called without the ticker being in the user's knowledge base — it's a read-only market data endpoint.

Response:

{
  "success": true,
  "ticker": "AAPL",
  "pe_history": [{"date": "2017-04", "pe": 16.2}, ...],
  "pb_history": [...],
  "current_pe_percentile": 0.82,
  "current_pb_percentile": 0.75
}

Response schema — full profile

{
  id, ticker, company_name, market,          // market = US | HK | CN
  current_price, pe_ratio, pb_ratio,
  pe_band_data,                              // 8yr history, same shape as /band endpoint
  financial_red_flags,                       // [{rule_id, severity: "high|med|low", message}]
  event_radar,                               // [{event_type, timestamp, headline}]
  ai_profile_summary,                        // markdown, ~500 chars
  status,                                    // "active" | "archived"
  last_viewed_at, last_updated_at, created_at
}

Typical Workflow

1. User: "Add NVDA to my research brain"
   → POST /api/research/profiles {"ticker": "NVDA"}
   → Present: "Added NVDA. Current PE 45, 2 red flags, PE at 85th percentile of 8yr range."

2. User: "What's in my knowledge base?"
   → GET /api/research/profiles
   → Present table: ticker · company · PE · last_updated · red flag count

3. User: "Show me my AAPL profile"
   → GET /api/research/profiles/AAPL
   → Present: summary, PE/PB bands, red flags list, event radar, linked thesis (if any)

4. User: "Refresh my TSLA profile"
   → POST /api/research/profiles/TSLA/refresh

Tier Limits

When a create hits the limit, the API returns upgrade_required: true. Surface this to the user with a prompt to upgrade at /pricing.

Output Formatting Tips

When presenting a profile to the user, highlight:

1. Ticker + company name and market flag (US / HK / CN)
2. Current price + PE / PB with percentile context ("PE 32, 85th percentile of 8yr range → rich")
3. Red flags — group by severity, show top 3
4. Event radar — most recent 3-5 events with dates
5. Linked thesis — if present, one-line buy reason + exit trigger summary
6. Staleness — if last_updated_at > 7d old, suggest a refresh

Related Skills

• alphagbm-investment-thesis — Attach buy thesis + exit triggers to a profile
• alphagbm-health-check — Detect stale / drifted profiles across the user's workspace
• alphagbm-stock-analysis — One-off deep analysis (not persisted to the knowledge base)
• alphagbm-theme-research — Group profiles into themes (AI infra, HK dividend, etc.)

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Duan Yongping Analysis

Instant Duan-style framing for a single ticker. Three scoped outputs:

1. Sell Put — if you're "willing to buy at $X", what does selling a Put at $X actually pay, and how does the cost basis work out if assigned?
2. Covered Call — if you hold 100 shares, what yield can you pick up by selling a ~5% OTM Call 25-50 DTE? If called away, what's the total return?
3. Panic Buy Context — what's VIX telling us right now? Are we at Duan's "extreme fear = extreme opportunity" tier (VIX ≥ 35), or is this a wait?

Each panel is tailored to the Duan framework: seller-only, rent-collection, holding quality companies indefinitely, and treating extreme VIX spikes as opportunity rather than threat.

Why This Is a Separate Skill

The generic alphagbm-options-strategy can compute any spread. But Duan Yongping's style has three very specific moves and a very specific philosophy. This skill packages that philosophy into a single call with Chinese-native copy that fits how Chinese retail investors actually talk about these trades.

How to Use

Input:

• ticker (required)
• buy_price (optional) — your "I'd happily buy at this price" level; defaults to spot × 0.95 if omitted

Output (each panel may be null if no suitable contract exists):

• sell_put: {strike, premium, annualized_yield_pct, if_assigned_cost_basis, delta, dte}
• covered_call: {strike, premium, annualized_yield_pct, upside_cap_pct, total_return_if_called_pct, dte}
• panic_buy: {vix, level, signal (bool), action_zh, action_en}
  • level ∈ normal / caution / extreme_fear
  • signal = true when VIX ≥ 35 (Duan-buy tier)

Plus meta: ticker, stock_price, expiry_date, dte, timestamp.

Example Queries

• Duan Yongping style AAPL — All three panels for Apple
• sell put NVDA willing to buy at 110 — Sell-Put sized for $110 entry
• covered call yield on TSLA — CC analysis at current price
• is VIX at Duan buy level — Panic-Buy panel alone (can also use alphagbm-vix-status)
• should I sell AAPL put at 180 — Sell-Put analysis at specific strike

Mock Data

Mock data in mock-data/duan-analysis/ — sample for AAPL with a 180 buy price.

API Endpoint

GET /api/options/duan-analysis?ticker={SYMBOL}&buy_price={PRICE}

Query params:

• ticker (required)
• buy_price (optional, float) — your preferred entry strike for Sell Put; defaults to spot × 0.95

Response shape:

{
  "success": true,
  "ticker": "AAPL",
  "stock_price": 185.4,
  "expiry_date": "2026-06-20",
  "dte": 41,
  "sell_put": {
    "strike": 180.0,
    "premium": 2.45,
    "annualized_yield_pct": 12.1,
    "implied_vol": 0.24,
    "delta": -0.28,
    "open_interest": 4821,
    "volume": 312,
    "if_assigned_cost_basis": 177.55,
    "dte": 41
  },
  "covered_call": {
    "strike": 195.0,
    "premium": 2.10,
    "annualized_yield_pct": 10.1,
    "implied_vol": 0.22,
    "delta": 0.32,
    "open_interest": 3200,
    "volume": 198,
    "upside_cap_pct": 5.18,
    "total_return_if_called_pct": 6.31,
    "dte": 41
  },
  "panic_buy": {
    "vix": 18.7,
    "level": "normal",
    "signal": false,
    "action_zh": "VIX 18.7 偏平静。段永平风格下此水位更适合卖 Put 等跌到心理价位，而不是主动抄底。",
    "action_en": "VIX 18.7 is calm. Duan-style strategy prefers Sell-Put \"waiting\" over proactive buying at this level."
  },
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: 1 option-analysis credit per call; 5-min cache per (ticker, buy_price) pair.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM FearScore

A weighted composite panic gauge, per ticker. Reconstructs the FearDesk framework in one API call: six orthogonal fear signals, each scored 0–100, then combined with fixed weights into a single number. Score ≥ 60 is the historical trigger for Bull Put Spread entries.

Scoring Weights

Each indicator has its own 0–100 sub-score with thresholds tuned so extreme readings contribute most. Missing inputs fall back to neutral values (and are flagged in components.*.fallback), so the endpoint never 500s on partial data.

Why It Exists

Most fear gauges are either VIX-only (miss per-ticker divergence) or opaque ("sentiment index: 72"). This breaks down exactly what drove the score so you can decide whether to trust it.

Backtest evidence: Across 146 live Bull Put Spread trades, entries at FearScore ≥ 60 delivered ~10.8% annualized ROC vs ~3.5% for unconditional entries — roughly 3× the alpha from a single filter. Use this as the market-timing layer on any premium-selling strategy.

How to Use

Input: A ticker symbol.

Output:

• fear_score — weighted total 0-100
• signal — boolean, true when fear_score ≥ threshold (default 60)
• threshold — current trigger value
• confidence — 0-1, fraction of the 6 indicators that used real (non-fallback) data
• components.{vix,iv_rank,rsi,volume_anomaly,pc_ratio,consecutive_down}:
  • value — raw input
  • score — 0-100 per-indicator score
  • weight — contribution weight
  • fallback — true if neutral default was used

Example Queries:

• fear score QQQ — Full breakdown of the 6 indicators for QQQ
• is NVDA oversold right now — RSI + FearScore composite
• BPS signal SPY — Check if entry threshold is hit
• when should I sell put AAPL — Timing via FearScore ≥ 60 rule
• how panicked is TSLA today — Per-ticker panic index with component breakdown
• why is QQQ fear score low — Component-by-component explanation

Mock Data

Mock data in mock-data/fear-score/ — example responses at neutral / elevated / signal-triggered readings.

API Endpoint

GET /api/options/fear-score?ticker={SYMBOL}

Query params:

• ticker (required) — stock symbol (US / HK / CN supported if whitelisted)

Response shape:

{
  "success": true,
  "ticker": "QQQ",
  "fear_score": 68.2,
  "signal": true,
  "threshold": 60,
  "confidence": 1.0,
  "components": {
    "vix": {"value": 28.4, "score": 82, "weight": 0.20, "fallback": false},
    "iv_rank": {"value": 78, "score": 78, "weight": 0.25, "fallback": false},
    "rsi": {"value": 24.1, "score": 88, "weight": 0.15, "fallback": false},
    "volume_anomaly": {"value": 2.3, "score": 72, "weight": 0.15, "fallback": false},
    "pc_ratio": {"value": 1.6, "score": 80, "weight": 0.15, "fallback": false},
    "consecutive_down": {"value": 3, "score": 60, "weight": 0.10, "fallback": false}
  },
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: 1 option-analysis credit per call; per-ticker 5-min cache (cache hits free).

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Greeks

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Provides a comprehensive Greeks dashboard for any single option contract or multi-leg position. Calculates first-order and second-order sensitivities, and generates scenario heatmaps showing how Greeks change as price and IV move.

Greeks Covered

Position-Level Analysis

For multi-leg positions, the skill aggregates Greeks across all legs and shows:

• Net Greeks: Total delta, gamma, theta, vega for the combined position
• Greeks per unit of capital: Normalized by margin requirement or net debit
• Risk concentration: Which leg contributes most to each Greek

API Endpoints

Greeks Calculator

Calculate Greeks for a single option from basic parameters:

POST /api/options/tools/greeks
Content-Type: application/json

{
  "spot": 150,
  "strike": 155,
  "expiry_days": 30,
  "iv": 0.25,
  "option_type": "call"
}

Parameters:

• spot (required): Current underlying price
• strike (required): Option strike price
• expiry_days (required): Days to expiration
• iv (required): Implied volatility as decimal (e.g., 0.25 for 25%)
• option_type (required): "call" or "put"

Implied Volatility Calculator

Reverse-solve for IV given market price:

POST /api/options/tools/implied-volatility
Content-Type: application/json

{
  "market_price": 4.50,
  "spot": 150,
  "strike": 155,
  "expiry_days": 30,
  "option_type": "call"
}

Parameters:

• market_price (required): Current market price of the option
• spot (required): Current underlying price
• strike (required): Option strike price
• expiry_days (required): Days to expiration
• option_type (required): "call" or "put"

How to Use

Input

• Required: Ticker + strike + expiry + type (for single contract), OR a position definition (list of legs)
• Optional: Underlying price override, IV override, date override (for forward-looking)

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "position": [
    {
      "leg": "AAPL 2026-04-18 220C",
      "quantity": 1,
      "greeks": {
        "delta": 0.52,
        "gamma": 0.035,
        "theta": -0.18,
        "vega": 0.32,
        "rho": 0.08,
        "charm": -0.003,
        "vanna": 0.012,
        "volga": 0.005
      }
    }
  ],
  "net_greeks": {
    "delta": 0.52,
    "gamma": 0.035,
    "theta": -0.18,
    "vega": 0.32,
    "rho": 0.08
  },
  "heatmap": {
    "price_axis": [200, 205, 210, 215, 220, 225, 230, 235],
    "iv_axis": [20, 25, 30, 35, 40],
    "delta_grid": [
      [0.12, 0.15, 0.20, 0.28, 0.38, 0.50, 0.62, 0.73],
      [0.14, 0.18, 0.24, 0.32, 0.42, 0.52, 0.63, 0.74],
      [0.16, 0.20, 0.27, 0.35, 0.45, 0.55, 0.65, 0.75],
      [0.18, 0.23, 0.30, 0.38, 0.48, 0.57, 0.67, 0.76],
      [0.20, 0.25, 0.32, 0.40, 0.50, 0.59, 0.68, 0.77]
    ],
    "pnl_grid": "..."
  },
  "insights": [
    "Position is net long delta (0.52) -- profits if stock rises",
    "Theta of -0.18 means $18/day time decay per contract",
    "Gamma of 0.035 means delta shifts ~3.5 for a $1 move"
  ]
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Greeks calculated from realistic option chain snapshots in mock-data/.

Related Skills

• alphagbm-options-score -- Greeks balance is a scoring factor for contract quality
• alphagbm-pnl-simulator -- Visualize how Greeks translate into actual P&L outcomes
• alphagbm-options-strategy -- See net Greeks for recommended strategies
• alphagbm-vol-surface -- Understand the IV inputs driving vega and vanna

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Knowledge Base Health Check

Periodic audit of a user's research workspace — stale profiles, drifted theses, orphan pages — with an overall 0-100 score and concrete recommendations.

When to use

• User asks for an overview of what's broken / out of date in their KB
• User wants to manually trigger a fresh health report
• User asks about stale profiles, thesis drift, or orphan items
• User mentions "健康体检" / "体检" / "audit" / "health check" / "什么需要更新"

Prerequisites

• API Key: env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override via ALPHAGBM_BASE_URL.
• Tier requirement for generate: POST /health/generate is Pro-only. Free/Plus users get 403 with upgrade_required: true. Free/Plus users can still read the latest auto-generated weekly report via GET /health.

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. Get latest health report

GET /api/research/health

Response when a report exists:

{
  "success": true,
  "has_report": true,
  "report_date": "2026-04-13",
  "overall_score": 78,
  "stale_profiles": [
    {"ticker": "AAPL", "days_since_update": 21}
  ],
  "thesis_drift": [
    {"ticker": "NVDA", "drift_reason": "Revenue growth now 12% vs 25% when thesis was written"}
  ],
  "orphan_pages": [
    {"ticker": "XYZ", "issue": "Profile has no thesis after 30 days"}
  ],
  "recommendations": [
    {"action": "refresh", "ticker": "AAPL", "reason": "21 days stale"},
    {"action": "review_thesis", "ticker": "NVDA", "reason": "growth decelerated"},
    {"action": "archive", "ticker": "XYZ", "reason": "orphan for 30+ days"}
  ],
  "created_at": "2026-04-13T02:00:00Z"
}

Response when no report yet:

{
  "success": true,
  "has_report": false
}

2. Generate fresh report (Pro-only)

POST /api/research/health/generate

Kicks off immediate audit across the user's profiles + theses + themes. Returns the new report in the same shape as GET.

Tier-blocked response (403):

{
  "success": false,
  "error": "Health check generation is a Pro feature.",
  "upgrade_required": true
}

Free/Plus users still get an auto-generated report weekly (served by GET), just can't trigger on-demand.

Response schema — report

{
  id, report_date,
  stale_profiles,          // [{ticker, days_since_update}]
  thesis_drift,            // [{ticker, drift_reason}]
  orphan_pages,            // [{ticker, issue}]
  overall_score,           // 0-100
  recommendations,         // [{action, ticker, reason}]
  created_at
}

Score band interpretation

Recommendation action types

• refresh — call POST /api/research/profiles/<ticker>/refresh
• review_thesis — surface the thesis + updated fundamentals; user decides to edit/close
• archive — DELETE /api/research/profiles/<ticker> or delete orphan
• create_thesis — profile exists but has no thesis; prompt user to write one

Skills that execute these actions: alphagbm-company-profile, alphagbm-investment-thesis.

Typical Workflow

1. User: "How's my research brain looking?"
   → GET /api/research/health
   → If has_report=false: "No report yet — your first auto-audit runs on <day>.
      Pro users can trigger one now."
   → If has_report=true: present score + top 3 recommendations

2. User (Pro): "Run a fresh health check now"
   → POST /api/research/health/generate
   → Present the new report

3. User (Free/Plus): "Run a health check"
   → POST /api/research/health/generate returns 403
   → Fall back: show the last weekly auto-report via GET, suggest upgrade

4. User: "Fix the stale ones"
   → For each recommendation with action=refresh:
     POST /api/research/profiles/<ticker>/refresh
   → Re-check: GET /api/research/health → show updated score

Output Formatting Tips

When presenting a health report:

1. Lead with overall score + band (color: green ≥ 75, yellow 60-74, red < 60)
2. Top 3 recommendations — most actionable first (refresh > review > archive)
3. Group by category — "3 stale profiles · 1 drifted thesis · 2 orphans" as a chip row
4. Each ticker actionable — pair the reason with a one-click (or one-prompt) action
5. Report date prominence — "as of 2026-04-13" so user knows freshness
6. If no report yet — explain the weekly cadence, offer Pro upgrade for on-demand

Related Skills

• alphagbm-company-profile — Run refresh on flagged stale profiles
• alphagbm-investment-thesis — Review / close drifted theses
• alphagbm-theme-research — Address orphan pages inside themes

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Hedge Advisor

"I own AAPL at $140 and it's now $180 — how do I protect the gains?"

Takes that question literally. Given a ticker + cost basis + position purpose, the skill classifies the holding into one of four scenarios and returns ready-to-trade hedge specs with strikes and costs already resolved from the live option chain.

Scenarios

What's Returned

For each recommendation spec, the skill resolves actual strikes and prices from the live option chain:

• Long Put: strike, DTE, cost_per_share, cost_per_contract, cost_pct_of_spot, delta, IV
• Collar: long_put_strike, short_call_strike, put_cost, call_credit, net_cost_per_share (negative = you receive a credit), breakeven analysis
• Tier-down / Position rules: static rules copy only

Also returns a position_rules[] array (single-name ≤20%, sector ≤30-35%, cash reserve 10-15%, etc.) for the normal-hold case.

How to Use

Input:

• ticker (required)
• cost_basis (required, float — your average entry price)
• purpose (optional, default long_term) — one of long_term / short_term / pre_earnings / just_bought

Output:

• Scenario label + reason (zh/en)
• Current price, cost basis, unrealized P&L %, recent drawdown %
• recommendations[] — each with type, priority, title, rationale, and resolved block containing the actual priced hedge
• position_rules[] — always-applicable sizing rules

Example Queries:

• hedge my AAPL at $140, now it's $180 → Gain Protection → Collar 95/110 quote
• I just bought NVDA at $110 on the dip, should I hedge? → Falling Knife or Bottom Fishing → Long Put 5% OTM 60-90 DTE
• how to protect my TSLA position → Gain Protection or Bottom Fishing based on PnL
• collar MSFT at cost 340 current 410 → Full collar pricing

Mock Data

Mock responses in mock-data/hedge-advisor/ — sample across all four scenarios.

API Endpoint

GET /api/options/hedge-advisor?ticker={SYMBOL}&cost_basis={PRICE}&purpose={PURPOSE}

Query params:

• ticker (required)
• cost_basis (required, float > 0)
• purpose (default long_term) — one of long_term / short_term / pre_earnings / just_bought

Response shape:

{
  "success": true,
  "ticker": "AAPL",
  "current_price": 180.0,
  "cost_basis": 140.0,
  "unrealized_pnl_pct": 28.57,
  "recent_drawdown_pct": 3.1,
  "purpose": "long_term",
  "scenario": {
    "scenario": "gain_protection",
    "label_zh": "浮盈怕坐电梯",
    "label_en": "Gain Protection",
    "reason_zh": "已浮盈 28.6%，需要保护已实现收益。",
    "reason_en": "Up 28.6% on cost — protect unrealized gains.",
    "unrealized_pnl_pct": 28.57
  },
  "recommendations": [
    {
      "type": "collar",
      "priority": 1,
      "title_zh": "Collar 95/110 锁定收益",
      "title_en": "Collar 95/110 lock-in",
      "rationale_zh": "...",
      "rationale_en": "...",
      "resolved": {
        "long_put_strike": 170.0,
        "short_call_strike": 200.0,
        "put_cost": 2.15,
        "call_credit": 2.45,
        "net_cost_per_share": -0.30,
        "net_cost_per_contract": -30,
        "is_credit": true,
        "dte": 62
      }
    },
    {"type": "tier_down", "priority": 2, ...}
  ],
  "position_rules": [
    {"rule_zh": "单票仓位 ≤ 20%", "rule_en": "Single ticker ≤20%", ...},
    ...
  ]
}

Pricing: 1 option-analysis credit per call; 5-min cache per (ticker, cost_basis, purpose).

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Investment Thesis

Turn "I bought this because…" into a tracked, monitored record. Each thesis pairs a prose buy-reason with structured sell conditions so the system can auto-detect when the reasoning no longer holds.

When to use

• User wants to document why they bought a stock
• User wants to set exit triggers (price, PE, fundamental breach)
• User asks which theses are still valid vs triggered
• User asks to update / refine an existing thesis
• User mentions "论据" / "买入理由" / "卖出条件" / "thesis" / "exit trigger"

Prerequisites

• API Key: env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override via ALPHAGBM_BASE_URL.
• Profile required: A thesis must attach to an existing company profile. If the user hasn't created a profile for the ticker, call POST /api/research/profiles first (see alphagbm-company-profile).

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. List theses

GET /api/research/theses?status=active

Response:

{
  "success": true,
  "theses": [
    { "id": 12, "ticker": "NVDA", "buy_thesis": "...", "status": "active", ... }
  ]
}

2. Get thesis by ticker

GET /api/research/theses/<TICKER>

Returns the active thesis for a ticker. 404 if none exists.

3. Create thesis

POST /api/research/theses
Content-Type: application/json

{
  "ticker": "NVDA",
  "buy_thesis": "AI capex cycle; data-center GPU moat; FCF > $60B.",
  "sell_conditions": [
    { "type": "price_drop_pct",  "value": 20 },
    { "type": "pe_above",        "value": 60 },
    { "type": "growth_below",    "value": 15 },
    { "type": "thesis_breach",   "value": "cloud capex guidance cut > 20%" }
  ]
}

Common sell_conditions types:

• price_drop_pct — drop from purchase/peak %
• pe_above / pb_above — valuation ceiling
• growth_below — revenue/earnings growth threshold
• thesis_breach — free-text qualitative trigger (monitored manually)

4. Update thesis (by id)

PUT /api/research/theses/<THESIS_ID>
Content-Type: application/json

{"buy_thesis": "updated prose", "sell_conditions": [...], "status": "closed"}

Partial updates allowed. Note: uses thesis_id (int), not ticker — read the id from a prior list or get.

5. Delete thesis (by id)

DELETE /api/research/theses/<THESIS_ID>

Hard-delete. Also uses numeric id.

Response schema — full thesis

{
  id, ticker,
  buy_thesis,                     // prose
  sell_conditions,                // [{type, value}]
  status,                         // "active" | "triggered" | "closed"
  thesis_score,                   // AI confidence 0-100 (if scored)
  ai_feedback,                    // AI critique of the thesis (markdown)
  triggered_at, trigger_detail,   // populated when status flips
  created_at, updated_at
}

Status lifecycle

active ──(sell condition fires)──▶ triggered
   │                                   │
   └────────(user closes)──▶ closed ◀──┘

When status = "triggered", trigger_detail shows which condition fired. Surface this to the user — it's the whole point of the system.

Typical Workflow

1. User: "I'm buying NVDA because AI capex is still accelerating"
   → (ensure profile exists — see alphagbm-company-profile)
   → POST /api/research/theses with buy_thesis + sell_conditions
   → Confirm: "Saved. Monitoring: price drop > 20%, PE > 60, growth < 15%."

2. User: "What are my active theses?"
   → GET /api/research/theses?status=active
   → Table: ticker · one-line thesis · conditions · score

3. User: "Any theses triggered?"
   → GET /api/research/theses?status=triggered
   → Alert list with trigger_detail explaining why

4. User: "Update my NVDA thesis — exit if PE > 70 instead of 60"
   → GET /api/research/theses/NVDA to find id
   → PUT /api/research/theses/<id> with revised sell_conditions

Output Formatting Tips

When presenting a thesis to the user, highlight:

1. Ticker + status (with color/emoji: active=green, triggered=red, closed=gray)
2. Buy thesis — first 2 sentences verbatim
3. Sell conditions — bulleted, human-phrased ("Exit if price drops 20%")
4. If triggered — which trigger fired, lead with that
5. AI feedback / score — if present, show as a pull-quote
6. Age — "written 3 weeks ago, reviewed 2 days ago"

Related Skills

• alphagbm-company-profile — Prerequisite. A thesis attaches to a profile.
• alphagbm-health-check — Surfaces theses that may have drifted from their original premise
• alphagbm-stock-analysis — Run a fresh analysis to sanity-check a thesis

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM IV Rank

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Calculates IV Rank and IV Percentile for any ticker, placing current implied volatility in historical context. Answers the key question: "Is IV high or low right now, and what should I do about it?"

Key Metrics

IV Zones and Trading Signals

API Endpoint

IV Snapshot (instant, no quota cost)

GET /api/options/snapshot/<SYMBOL>

Returns: ATM IV, IV Rank, HV 30d, VRP, VRP level. This endpoint is free and does not count against your analysis quota.

Volatility Risk Premium (VRP)

VRP = Implied Vol - Historical Vol

VRP measures the gap between what the market expects (IV) and what actually happens (HV). It is a key signal for whether to sell or buy premium.

How to Use

Input

• Required: Ticker symbol
• Optional: Lookback period (default 252 days), IV measure (30-day ATM, 60-day, or custom)

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "iv_current": 28.5,
  "iv_rank": 42,
  "iv_percentile": 55,
  "iv_52w_high": 48.2,
  "iv_52w_low": 18.8,
  "iv_52w_mean": 30.1,
  "hv_30d": 25.2,
  "hv_iv_ratio": 0.88,
  "zone": "moderate",
  "signal": "No strong IV edge. Use directional conviction to choose strategy.",
  "iv_history": {
    "dates": ["2025-04-01", "2025-04-02", "..."],
    "iv_values": [32.1, 31.8, "..."],
    "hv_values": [28.5, 28.3, "..."]
  },
  "notable_events": [
    {"date": "2026-01-28", "iv": 48.2, "event": "Earnings spike"},
    {"date": "2025-08-05", "iv": 44.1, "event": "Market selloff"}
  ]
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. IV history uses realistic 252-day data from mock-data/.

Related Skills

• alphagbm-vol-surface -- Full 3D IV landscape across strikes and expirations
• alphagbm-vol-smile -- IV skew for a specific expiration
• alphagbm-options-strategy -- IV zone informs whether to buy or sell premium
• alphagbm-options-score -- IV attractiveness is a key scoring factor

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Macro View

Track key macro indicators — VIX, US10Y, DXY, gold, oil, BTC, etc. — in the user's knowledge base. Each indicator comes with auto-computed impact analysis linked to the user's actual holdings.

When to use

• User wants to start tracking a macro variable (VIX, yields, USD, gold…)
• User asks for current macro dashboard / snapshot
• User asks how a macro change affects their portfolio
• User wants to stop tracking an indicator
• User mentions "宏观" / "VIX" / "美债" / "美元" / "macro" / "yield"

Prerequisites

• API Key: env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override via ALPHAGBM_BASE_URL.
• No profile requirement — macro tracking is independent of the company profile list.

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. List tracked indicators (also returns supported catalog)

GET /api/research/macro

Response:

{
  "success": true,
  "indicators": [
    {
      "indicator_key": "VIX",
      "display_name": "CBOE Volatility Index",
      "current_value": 18.2,
      "previous_value": 16.8,
      "change_pct": 8.3,
      "impact_analysis": "Rising VIX — elevated uncertainty. Your NVDA & TSLA positions are high-beta; consider…",
      "last_updated_at": "2026-04-13T10:15:00Z"
    }
  ],
  "supported": {
    "VIX":    {"name": "CBOE Volatility Index",   "category": "volatility"},
    "US10Y":  {"name": "US 10-Year Treasury",     "category": "yields"},
    "DXY":    {"name": "US Dollar Index",         "category": "currency"},
    "GOLD":   {"name": "Gold Spot",               "category": "commodity"},
    ...
  }
}

The supported field is the catalog of valid indicator_key values. Use it to present options when the user asks "what can I track".

2. Add indicator

POST /api/research/macro
Content-Type: application/json

{"indicator_key": "VIX"}

400 response for unsupported keys:

{
  "success": false,
  "error": "Unsupported indicator. Supported: ['VIX', 'US10Y', 'DXY', ...]"
}

3. Remove indicator

DELETE /api/research/macro/<INDICATOR_KEY>

Uses the key (VIX, not an id). 404 if not tracked.

Response schema — indicator

{
  id, indicator_key,
  display_name,              // human-readable name
  current_value,             // most recent reading
  previous_value,            // for change_pct computation
  change_pct,                // % change
  impact_analysis,           // AI-generated, references user's holdings
  last_updated_at
}

Common indicator keys

Always call GET /api/research/macro first to fetch the live supported catalog — keys may be added/retired.

Typical Workflow

1. User: "Track VIX and the 10-year yield"
   → POST /api/research/macro {"indicator_key": "VIX"}
   → POST /api/research/macro {"indicator_key": "US10Y"}
   → Confirm both added with their current values

2. User: "What's the macro picture?"
   → GET /api/research/macro
   → Present each indicator: value, change, impact on their holdings

3. User: "Stop tracking DXY"
   → DELETE /api/research/macro/DXY

4. User: "Is high VIX hurting my positions?"
   → GET /api/research/macro → read VIX's impact_analysis field
   → The impact_analysis already references the user's specific holdings

Output Formatting Tips

When presenting macro indicators:

1. Table-first for multi-indicator views: key · value · change% · one-line impact
2. Highlight change direction with arrow/color (↑ red for VIX/yields up, ↓ green etc.)
3. Lead with impact_analysis when user asks "how does X affect my portfolio" — it's pre-computed with their holdings in mind
4. Stale data — if last_updated_at > 1d old, note "data may be stale, refresh coming"
5. When user asks "what can I track", show the supported catalog grouped by category

Related Skills

• alphagbm-company-profile — Macro impact analysis references the user's profiles
• alphagbm-market-sentiment — Broader cross-asset sentiment read
• alphagbm-iv-rank — Options-specific volatility context

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Howard Marks Cycle

"Cycles are real — the shape just isn't predictable." Howard Marks's framework rejects forecasting and replaces it with cycle-position awareness: offense when others are pessimistic, defense when others are optimistic.

This skill gives you the one number Marks's entire philosophy implies: where are we right now.

The Cycle Score

Each signal is mapped to its own cycle component 0-100, then weighted:

Weights renormalize when data points are missing (e.g., P/C not available).

Posture Bands

• 0-24 → OFFENSE_HARD — extreme fear is opportunity. Buy aggressively.
• 25-39 → OFFENSE — add, sell vol (short premium).
• 40-59 → NEUTRAL — maintain positions, watch for shifts.
• 60-74 → DEFENSE — don't add, brace for volatility.
• 75-100 → DEFENSE_HARD — trim, buy protection (long puts / collars).

Why This Is a Separate Skill

alphagbm-vix-status gives just a VIX tier. alphagbm-market-sentiment gives a sentiment dashboard. This skill is the one-call Marks-specific read: "given everything I know about sentiment + valuation, what's the posture?"

How to Use

Input: none (market-level, no ticker)

Output:

• cycle_score: integer 0-100
• posture: one of OFFENSE_HARD / OFFENSE / NEUTRAL / DEFENSE / DEFENSE_HARD
• posture_zh, posture_en: natural-language prescription
• components: per-signal {value, cycle_component} breakdown

Example Queries

• where are we in the cycle right now → headline cycle number + posture
• should I be playing offense or defense → posture directly answers
• Howard Marks read on the market → same data, framed as Marks would
• is this a buying cycle → cycle < 30 → yes; cycle > 60 → no
• current sentiment across VIX and IV rank → components breakdown

Mock Data

Mock data in mock-data/marks-cycle/ — sample showing NEUTRAL position.

API Endpoint

GET /api/masters/marks-cycle

No body, no auth required.

Response shape:

{
  "success": true,
  "cycle_score": 47,
  "posture": "NEUTRAL",
  "posture_zh": "中性 — 维持既定仓位,观察情绪变化",
  "posture_en": "Neutral — maintain positions, watch sentiment",
  "components": {
    "vix": {"value": 22.5, "cycle_component": 48},
    "iv_rank": {"value": 55, "cycle_component": 45}
  },
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: free — no auth, no credit deduction. 5-min cache.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Options Strategy

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Given a market view and a ticker, recommends the best multi-leg option strategies ranked by risk/reward profile. Selects optimal strikes and expirations automatically using AlphaGBM's scoring engine.

Four Core Strategies and Trend Alignment

Trend alignment scoring: The scoring model rewards contracts that match the prevailing trend. For Sell Put, a downtrend scores 100 (counter-intuitive: you want to sell puts into weakness for higher premium), while an uptrend scores 30. For Buy Call, bullish momentum is weighted at 25%.

Supported Strategy Templates (15+)

Risk-Return Profiles

Strategy Selection Logic

1. Match user's market view to candidate strategies
2. Filter by IV environment (high IV favors selling premium; low IV favors buying)
3. Score each candidate using risk/reward, probability of profit, and capital efficiency
4. Rank and return the top 3 recommendations with full details

API Endpoints

Strategy Templates

List all available strategy templates:

GET /api/options/tools/strategy/templates

Strategy Builder

Build a strategy from a template with specific parameters:

POST /api/options/tools/strategy/build
Content-Type: application/json

{
  "mode": "template",
  "template_id": "bull_call_spread",
  "spot": 150.0,
  "expiry_days": 30,
  "strikes": [140, 145, 150, 155, 160]
}

Options Scanner

Scan across tickers for strategies matching your criteria:

POST /api/options/tools/scan
Content-Type: application/json

{
  "strategies": ["covered_call", "cash_secured_put"],
  "tickers": ["AAPL", "NVDA"],
  "min_yield_pct": 1.0
}

How to Use

Input

• Required: Ticker symbol + market view (bullish / bearish / neutral / volatile)
• Optional: Max capital, target expiration, risk tolerance (conservative / moderate / aggressive)

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "market_view": "bullish",
  "iv_environment": "moderate",
  "recommendations": [
    {
      "strategy": "Bull Call Spread",
      "rank": 1,
      "score": 8.5,
      "legs": [
        {"action": "buy", "type": "call", "strike": 215, "expiry": "2026-04-18", "price": 7.20},
        {"action": "sell", "type": "call", "strike": 225, "expiry": "2026-04-18", "price": 3.40}
      ],
      "max_profit": 620,
      "max_loss": 380,
      "breakeven": [218.80],
      "probability_of_profit": 0.58,
      "risk_reward_ratio": 1.63,
      "net_debit": 380,
      "greeks": {
        "delta": 0.32,
        "gamma": 0.012,
        "theta": -0.08,
        "vega": 0.14
      },
      "rationale": "Moderate bullish exposure with capped risk. IV is fair -- debit spread preferred over naked call."
    }
  ]
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Strategy recommendations use realistic chain data from mock-data/.

Related Skills

• alphagbm-options-score -- Scores the individual contracts used in each leg
• alphagbm-pnl-simulator -- Simulate P&L over time for any recommended strategy
• alphagbm-greeks -- Deep-dive into position Greeks for the chosen strategy
• alphagbm-iv-rank -- Check if IV environment favors buying or selling premium

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM P&L Simulator

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Simulates profit and loss for any option position across multiple dimensions -- underlying price, implied volatility, and time to expiration. Produces P&L diagrams, breakeven analysis, and probability-weighted outcome distributions.

Four Core Strategies for Context

Simulation Capabilities

Supported Position Types

• Single leg (long call, long put, short call, short put)
• Two-leg spreads (vertical, calendar, diagonal)
• Three-leg combinations (butterflies, ratio spreads)
• Four-leg combinations (iron condors, iron butterflies, double diagonals)
• Arbitrary multi-leg custom positions

API Endpoint

P&L Simulator

POST /api/options/tools/simulate
Content-Type: application/json

{
  "symbol": "AAPL",
  "spot": 150.0,
  "legs": [
    {"action": "buy", "option_type": "call", "strike": 145, "expiry_days": 30, "iv": 0.26},
    {"action": "sell", "option_type": "call", "strike": 150, "expiry_days": 30, "iv": 0.25}
  ]
}

Parameters:

• symbol (required): Ticker symbol
• spot (required): Current underlying price
• legs (required): Array of option legs, each with:
  • action: "buy" or "sell"
  • option_type: "call" or "put"
  • strike: Strike price
  • expiry_days: Days to expiration
  • iv: Implied volatility as decimal (e.g., 0.26 for 26%)

How to Use

Input

• Required: Position definition (legs with strike, expiry, type, quantity, entry price)
• Optional: Scenario parameters (price range, IV shift, target date), number of Monte Carlo paths

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "position": {
    "strategy": "Bull Call Spread",
    "legs": [
      {"action": "buy", "type": "call", "strike": 215, "expiry": "2026-04-18", "price": 7.20, "qty": 1},
      {"action": "sell", "type": "call", "strike": 225, "expiry": "2026-04-18", "price": 3.40, "qty": 1}
    ],
    "net_debit": 380
  },
  "pnl_at_expiry": {
    "price_axis": [195, 200, 205, 210, 215, 218.8, 220, 225, 230, 235],
    "pnl_axis":   [-380, -380, -380, -380, -380, 0, 120, 620, 620, 620]
  },
  "pnl_over_time": {
    "dates": ["2026-03-29", "2026-04-04", "2026-04-11", "2026-04-18"],
    "curves": {
      "at_210": [-180, -220, -290, -380],
      "at_218": [50, 30, 10, -20],
      "at_225": [320, 400, 510, 620]
    }
  },
  "breakevens": [218.80],
  "max_profit": 620,
  "max_loss": 380,
  "risk_reward_ratio": 1.63,
  "probability_of_profit": 0.56,
  "expected_value": 42.50,
  "scenarios": {
    "price_down_10pct": {"pnl": -380, "pnl_pct": -100},
    "price_up_10pct": {"pnl": 620, "pnl_pct": 163},
    "iv_crush_50pct": {"pnl": -85, "note": "IV drop hurts long spread slightly"},
    "iv_spike_50pct": {"pnl": 120, "note": "IV rise helps long spread slightly"}
  }
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Simulations use realistic pricing models calibrated to mock-data/ snapshots.

Related Skills

• alphagbm-options-strategy -- Get strategy recommendations, then simulate them here
• alphagbm-greeks -- Understand the Greeks driving the P&L changes
• alphagbm-iv-rank -- Context for whether IV scenarios are realistic
• alphagbm-vol-surface -- Full IV landscape for calibrating simulations

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Take-Profit Strategy Lab

Answer one question mechanically for any ticker: Can you just hold it, or do you need to actively take profits? Most retail losses come from poor exits, not poor entries. This skill quantifies the exit decision with 10 years of daily data.

The Core Metric: Rollercoaster Rate

A "rollercoaster event" happens when an entry's paper profit exceeds +50% and then falls more than 50% from that peak before exit. Example: enter at 100, peak at 190, fall back to 90 — you didn't lose money, but the 90 of peak profit you "touched" evaporated, and the journey was brutal.

Rollercoaster rate varies up to 97 percentage points across instruments:

• Broad-index ETFs (SPY, VTI): 0% — hold forever
• Blue chips (AAPL, MSFT): 0% — hold forever
• Sector ETFs (SOXX, XLK): 0% — hold forever
• Large-cap mega-caps (META, AMZN): ~47% — tiered exit preferred
• HK tech (腾讯, 阿里): ~49% — tiered exit preferred
• High growth (NVDA, TSLA, AMD): ~85% — tiered exit mandatory
• Crypto-related (COIN, MSTR): ~90% — tiered exit mandatory
• Leveraged ETFs (TQQQ, SOXL): ~97% — structurally un-holdable

Whether you can hold is an instrument property, not an attitude problem.

Strategy Universe (15 total)

• A family (sell all at trigger): A_+50%, A_+100%, A_+200%
• B family (tiered): B_50/100/200 (default), B_30/60/100, B2_20/40/80, B3_40/80/150, B5 back-weighted, B6 front-weighted
• C_10x (conviction hold)
• D (-20% / -30% trailing stop) — loses to hold on every tested ticker
• E (never sell / long-hold)
• F (peak-pullback after +50% activation)
• G (HV-aware: picks A_+100% or A_+200% based on entry-day vol)

How to Use

Input:

• ticker (required) — any US / HK / CN stock, ETF, or leveraged ETF

Output:

• Profile: color (green/amber/red) + special_flag (no_hold for leveraged ETFs, reverse_alpha for declining stocks where active selling beats hold)
• Headline numbers: rollercoaster_rate, max_drawdown, hold_cagr
• strategy_results: 15 strategies, each with {cagr, rc, mdd} medians
• Provenance: sample_size (typically ~120 entry points), period, computed_at

The caller is expected to:

1. Display the headline profile
2. Recommend a strategy matching user's personality + position size (front-end logic)
3. Generate concrete GTC limit-sell orders at entry × 1.5 / 2.0 / 3.0 etc.

Example Queries

• should I hold TQQQ long-term → no_hold flag + rollercoaster 97% → tiered exit
• take-profit strategy for NVDA → high-growth profile, B_50/100/200 default
• is AAPL holdable → blue-chip profile, 0% rollercoaster, hold recommended
• when should I sell COIN → crypto profile, mandatory tiered exit
• rollercoaster rate for SPY → 0%, long-hold optimal
• backtest sell strategies for MSFT → full 15-strategy comparison

Mock Data

Mock data in mock-data/take-profit/ — sample responses for TQQQ (no_hold), AAPL (hold-optimal), and PYPL (reverse_alpha).

API Endpoint

POST /api/stock/take-profit-analyze
Content-Type: application/json

Request body:

{"ticker": "TQQQ"}

Also available for reading the cached library (no quota):

GET /api/stock/take-profit-library

Returns list of already-cached tickers with their headline numbers — useful for agents to know which queries are instant vs first-time.

Response shape:

{
  "success": true,
  "ticker": "TQQQ",
  "color": "red",
  "special_flag": "no_hold",
  "rollercoaster_rate": 97,
  "max_drawdown": -82,
  "hold_cagr": 37.0,
  "strategy_results": {
    "A_50": {"cagr": 6.0, "rc": 21, "mdd": -32},
    "A_100": {"cagr": 11.5, "rc": 44, "mdd": -50},
    "A_200": {"cagr": 17.6, "rc": 64, "mdd": -62},
    "B_50_100_200": {"cagr": 12.0, "rc": 36, "mdd": -42},
    "B6_front": {"cagr": 11.0, "rc": 29, "mdd": -37},
    "E_hold": {"cagr": 37.0, "rc": 97, "mdd": -82},
    "...": "..."
  },
  "sample_size": 120,
  "period": {"start": "2014-04-20", "end": "2026-04-20"},
  "computed_at": "2026-04-24T08:00:00"
}

Pricing: 1 stock-analysis credit per first-time ticker compute; DB-cached for 30 days globally — once computed, all users get instant reads (including cache hits within 5 min in-process). Cache hits do not deduct credits.

First-time compute takes ~30s (10 years of daily data × 15 strategies × ~120 entry points = ~1800 simulations). Subsequent reads are <100 ms.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Tepper Panic-Buy Signal

David Tepper made two historic panic-bottom calls: March 2009 ("I'm betting on the Fed") and March 2020 (COVID bottom). Both had identical fingerprints: VIX spiked > 40, multi-indicator fear was maxed, and Tepper loaded quality large-caps (banks 2009, SPY/QQQ 2020).

This skill mechanizes that fingerprint.

The Signal Logic

Three gates must all pass for the signal to arm:

1. VIX ≥ 35 — extreme fear, not just elevated
2. FearScore ≥ 80 — multi-indicator panic (reuses existing FearScore module: VIX + IV Rank + RSI + Volume + Put/Call + Consec-Down-Days)
3. Quality filter — market cap > $50B AND profit margin > 0 (no memes, concept stocks, or pre-revenue small caps)

When all three pass → signal: true, level: armed.

Signal Levels

Tepper's own framework includes "do nothing" as a first-class state. Most calls to this endpoint will return cold — that's by design. The value isn't in the signal firing often, it's in never missing a VIX > 40 event.

Why This Is a Separate Skill

alphagbm-fear-score gives the raw panic index. alphagbm-vix-status gives the VIX tier. This skill combines them with Tepper's specific criteria (quality filter + threshold rules) to produce a single yes/no decision.

How to Use

Input:

• ticker (optional, default SPY) — the quality-filter applies to this ticker

Output:

• vix, fear_score — the two input signals
• vix_pass, fear_pass, quality_pass — per-gate booleans
• signal — final boolean
• level — armed / watch / near / cold
• recommended_etfs — ["SPY", "QQQ", "DIA"] (quality large-cap universe)
• advice_zh, advice_en — natural-language prescription

Example Queries

• is this a Tepper buy signal → call with default SPY
• panic-buy check on QQQ → substitute QQQ
• should I buy the panic now → returns cold if calm → "wait, this is the patience state"
• am I missing a historic bottom → the only time this fires, the answer is "yes, don't miss it"
• Tepper-style entry for DIA → quality large-cap → passes quality filter

Mock Data

Mock data in mock-data/tepper-signal/ — samples for armed (VIX 42, Fear 85) and cold (VIX 17, Fear 35).

API Endpoint

POST /api/masters/tepper-signal
Content-Type: application/json

Request body:

{"ticker": "SPY"}

Response shape:

{
  "success": true,
  "ticker": "SPY",
  "vix": 42.0,
  "fear_score": 85,
  "vix_pass": true,
  "fear_pass": true,
  "quality_pass": true,
  "signal": true,
  "level": "armed",
  "recommended_etfs": ["SPY", "QQQ", "DIA"],
  "advice_zh": "信号激活 — VIX 42.0、FearScore 85。按 Tepper 2009/2020 规则,分批买入大盘质量 ETF (SPY, QQQ, DIA)。不要买概念股或小盘股。",
  "advice_en": "Signal armed — VIX 42.0, FearScore 85. Per Tepper 2009/2020, scale into quality large-cap ETFs (SPY, QQQ, DIA). No memes, no small caps.",
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: 1 option-analysis credit per call; 5-min cache per ticker (cache hits free).

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Theme Research

Group related tickers into named investment themes with an AI-generated summary and news keyword watchlist. Each theme is a lightweight basket you can track at the concept level.

When to use

• User wants to organize tickers by theme (AI infra, HK dividend, EV supply chain, biotech…)
• User asks to view a specific theme's holdings + latest summary
• User wants to add or remove tickers from a theme
• User wants the system to monitor news around a topic
• User mentions "主题" / "theme" / "basket" / "篮子" / "板块"

Prerequisites

• API Key: env ALPHAGBM_API_KEY (format agbm_xxxx…).
• Base URL: default https://alphagbm.zeabur.app. Override via ALPHAGBM_BASE_URL.
• Tier limits apply: Free tier is capped on themes — check_profile_limit mirrors the profile limit model. Check limits.max_themes via the dashboard endpoint.

API Endpoints

All endpoints require Authorization: Bearer $ALPHAGBM_API_KEY.

1. List themes

GET /api/research/themes

Response:

{
  "success": true,
  "themes": [
    {
      "id": 7,
      "theme_name": "AI Infrastructure",
      "description": "Picks & shovels for the AI capex cycle",
      "tickers": ["NVDA", "AVGO", "MSFT", "ORCL"],
      "news_keywords": ["AI capex", "data center", "hyperscaler"],
      "theme_summary": "Capex guidance up across 4 hyperscalers...",
      "last_updated_at": "2026-04-13T09:00:00Z"
    }
  ]
}

2. Get theme detail (aggregated)

GET /api/research/themes/<THEME_ID>

Returns the theme + aggregated data across its tickers (average price change, top movers, recent news matching keywords). 404 if not found or not owned.

3. Create theme

POST /api/research/themes
Content-Type: application/json

{
  "theme_name": "AI Infrastructure",
  "description": "Picks & shovels for AI capex",
  "tickers": ["NVDA", "AVGO", "MSFT"],
  "news_keywords": ["AI capex", "data center"]
}

4. Update theme (by id)

PUT /api/research/themes/<THEME_ID>
Content-Type: application/json

{"tickers": ["NVDA", "AVGO", "MSFT", "ORCL"], "news_keywords": [...]}

Partial update. Any of the fields from create are accepted.

5. Delete theme (by id)

DELETE /api/research/themes/<THEME_ID>

Hard-delete. Doesn't affect the underlying company profiles.

Response schema — theme

{
  id, theme_name, description,
  tickers,                  // array of ticker strings
  news_keywords,            // array of phrases for news matching
  theme_summary,            // AI-generated narrative (markdown)
  last_updated_at, created_at
}

Theme detail endpoint (GET /themes/<id>) additionally includes aggregated fields like top movers and recent matched news — the exact shape is service-side and stable for display, not for programmatic parsing.

Typical Workflow

1. User: "Create an AI infra theme with NVDA, AVGO, MSFT"
   → POST /api/research/themes
     {"theme_name": "AI Infrastructure", "tickers": ["NVDA","AVGO","MSFT"],
      "news_keywords": ["AI capex", "data center"]}
   → Confirm theme created; mention it'll start accumulating summary + news

2. User: "What themes do I have?"
   → GET /api/research/themes
   → Table: theme · ticker count · last updated · summary excerpt

3. User: "Add ORCL to my AI theme"
   → GET /api/research/themes (find id)
   → PUT /api/research/themes/<id> {"tickers": [... + "ORCL"]}

4. User: "What's happening in my HK dividend theme?"
   → GET /api/research/themes/<id>
   → Lead with theme_summary + aggregated movers + matched news

Output Formatting Tips

When presenting themes:

1. List view — theme name · ticker count · "updated Xd ago" · 1-sentence summary
2. Detail view — lead with theme_summary (AI narrative), then ticker grid with % change, then recent matched news
3. Keyword hygiene — if the user creates a theme with no news_keywords, prompt: "Want me to watch for any news phrases? E.g., 'AI capex', 'hyperscaler'"
4. Ticker overlap — when creating a new theme, check if tickers already exist in other themes; it's fine (tickers can be in multiple themes) but worth mentioning

Related Skills

• alphagbm-company-profile — Themes reference profiles; creating a theme with untracked tickers still works but they won't have profile data
• alphagbm-health-check — Flags orphan tickers that are in themes but no longer in any profile
• alphagbm-compare — Side-by-side comparison for tickers within a theme

Powered by AlphaGBM — Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM VIX Status

The single VIX number, translated into the 5-tier framework that options sellers actually use to size up and down. Use this as the market-wide backdrop before any per-ticker analysis.

What This Skill Does

Maps the raw VIX value to one of five strategy zones:

Also returns:

• mean_1y — 1-year mean VIX for comparison
• percentile_1y — where today's VIX sits in last year's distribution
• distribution_1y_pct — % of last year spent in each of the 5 tiers

How to Use

Input: No parameters — market-wide indicator.

Output:

• vix — current close
• level — one of calm / normal / sweet_spot / caution / extreme_fear
• color — one of blue / green / yellow / orange / red
• label — zh/en short name (e.g. 卖方甜蜜区 / Seller Sweet Spot)
• strategy_hint — zh/en actionable guidance paragraph
• percentile_1y, mean_1y
• distribution_1y_pct — fraction of the past year in each tier

Example Queries:

• what's VIX right now — Current level + tier + strategy hint
• is this a good time for BPS — Check if VIX is in the "sweet spot" (20–25)
• should I sell premium today — Seller-view classification
• market fear gauge — VIX with contextual interpretation
• how often has VIX been in extreme fear this year — 1-year distribution

Mock Data

Mock data in mock-data/vix-status/ — sample responses across the 5 tiers.

API Endpoint

GET /api/options/vix-status

No parameters. Returns:

{
  "success": true,
  "vix": 22.5,
  "mean_1y": 18.3,
  "percentile_1y": 68.5,
  "level": "sweet_spot",
  "color": "yellow",
  "label": {"zh": "卖方甜蜜区", "en": "Seller Sweet Spot"},
  "strategy_hint": {"zh": "BPS 权利金变肥，积极开仓", "en": "BPS premiums get juicy — actively open positions"},
  "distribution_1y_pct": {"calm": 12.5, "normal": 45.2, "sweet_spot": 28.7, "caution": 11.3, "extreme_fear": 2.3},
  "timestamp": "2026-04-24T08:00:00"
}

Pricing: free (no quota deduction). 5-minute server-side cache.

Related Skills

Powered by AlphaGBM — Real-data options & research intelligence. 10K+ users.

AlphaGBM Volatility Smile

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Analyzes the volatility smile (or skew) for a single expiration -- the curve of implied volatility plotted against strike prices. Reveals how the market prices tail risk, directional fear, and supply/demand imbalances across the options chain.

Key Outputs

What Smile Shape Means for Trading

API Endpoint

Volatility Smile

GET /api/options/tools/vol-smile/<SYMBOL>?expiry=2026-04-17

Query parameters:

• expiry (optional): Expiration date in YYYY-MM-DD format. Defaults to nearest monthly expiry if omitted.

Returns the smile curve (strikes, IVs, deltas), skew metrics, shape classification, and skew percentile for the specified expiration.

How to Use

Input

• Required: Ticker symbol
• Optional: Expiration date (defaults to nearest monthly), moneyness range

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "expiry": "2026-04-18",
  "dte": 20,
  "smile": {
    "strikes": [190, 195, 200, 205, 210, 215, 220, 225, 230, 235, 240],
    "ivs":    [42.1, 39.5, 36.8, 34.0, 31.5, 29.2, 27.5, 28.8, 30.5, 32.8, 35.2],
    "deltas": [-0.10, -0.15, -0.22, -0.30, -0.40, -0.48, 0.52, 0.42, 0.32, 0.22, 0.14]
  },
  "skew_metrics": {
    "skew_25d": -8.3,
    "risk_reversal_25d": -2.45,
    "skew_10d": -14.6,
    "atm_iv": 28.3
  },
  "shape": "normal",
  "skew_percentile": 72,
  "interpretation": "Put skew is moderately steep (72nd percentile). OTM puts are pricing ~8 vol points above equidistant calls -- standard hedging demand with slight elevation."
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Smile data uses realistic IV snapshots from mock-data/.

Related Skills

• alphagbm-vol-surface -- See the full 3D surface across all expirations
• alphagbm-iv-rank -- Is overall IV high or low vs. history?
• alphagbm-options-strategy -- Steep skew suggests certain spread strategies
• alphagbm-options-score -- Use skew insights to find better-scored contracts

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.

AlphaGBM Volatility Surface

Prerequisites

• API Key: Set env ALPHAGBM_API_KEY (format agbm_xxxx...).
• Base URL: Default https://alphagbm.zeabur.app. Override with env ALPHAGBM_BASE_URL.

What This Skill Does

Builds a 3D volatility surface for any optionable ticker, mapping implied volatility across two dimensions -- strike price (moneyness) and time to expiration. Identifies where options are cheap, expensive, or anomalous relative to the surface.

Key Outputs

What the Surface Tells You

• Contango (front IV < back IV): Normal market, no near-term fear
• Backwardation (front IV > back IV): Near-term event expected (earnings, FDA, etc.)
• Steep skew: Market pricing tail risk in puts -- hedging demand is high
• Flat skew: Balanced sentiment -- no strong directional fear
• Anomaly detected: A specific contract is mispriced vs. neighbors -- potential opportunity

Volatility Risk Premium (VRP)

VRP = Implied Vol - Historical Vol

API Endpoints

Volatility Surface (3D)

GET /api/options/tools/vol-surface/<SYMBOL>

Returns the full 3D volatility surface with moneyness axis, expiry axis, and IV grid.

IV Snapshot (quick check, no quota cost)

For a fast ATM IV / IV Rank / HV / VRP check without pulling the full surface:

GET /api/options/snapshot/<SYMBOL>

Returns: ATM IV, IV Rank, HV 30d, VRP, VRP level.

How to Use

Input

• Required: Ticker symbol
• Optional: Moneyness range (e.g., 0.8-1.2), expiration range (e.g., 7-90 days)

Output Structure

{
  "ticker": "AAPL",
  "price": 218.45,
  "surface": {
    "moneyness_axis": [0.85, 0.90, 0.95, 1.00, 1.05, 1.10, 1.15],
    "expiry_axis": ["2026-04-04", "2026-04-18", "2026-05-16", "2026-06-20"],
    "iv_grid": [
      [38.2, 34.5, 31.0, 28.5, 30.2, 33.1, 36.8],
      [36.1, 33.0, 29.8, 27.2, 28.9, 31.5, 34.9],
      [34.5, 31.8, 28.5, 26.0, 27.5, 30.0, 33.2],
      [33.0, 30.5, 27.8, 25.5, 26.8, 29.0, 31.8]
    ]
  },
  "atm_term_structure": {
    "2026-04-04": 28.5,
    "2026-04-18": 27.2,
    "2026-05-16": 26.0,
    "2026-06-20": 25.5
  },
  "skew": {
    "2026-04-18": {"25d_put_iv": 33.0, "25d_call_iv": 28.9, "skew": -4.1}
  },
  "shape": "contango",
  "anomalies": [
    {
      "strike": 200,
      "expiry": "2026-04-18",
      "iv": 38.5,
      "expected_iv": 34.2,
      "deviation_sigma": 2.3,
      "signal": "overpriced"
    }
  ]
}

Example Queries

Mock Data

Demo tickers available without API key: AAPL, NVDA, SPY, TSLA, META. Surface data uses realistic IV snapshots from mock-data/.

Related Skills

• alphagbm-vol-smile -- Zoom into a single expiration's smile/skew curve
• alphagbm-iv-rank -- Is IV high or low vs. its own history?
• alphagbm-options-score -- Use surface insights to find the best-scored contracts
• alphagbm-options-strategy -- High IV surface suggests selling premium; low IV suggests buying

Powered by AlphaGBM -- Real-data options & research intelligence for traders and AI agents. 10K+ users.