Skip to content

mba811/skills

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AlphaGBM Skills

See what options are pricing in — with real data, not guesswork.

26 AI skills for options & research intelligence · Built on real market data · Trusted by 10,000+ traders

MIT License Skills Users

Website · Documentation · Quick Start · Contributing


AlphaGBM options analysis output

30-Second Demo

git clone https://github.com/AlphaGBM/skills.git .claude/skills/alphagbm

Then ask your AI: "Analyze AAPL options using AlphaGBM" — works instantly with built-in data, no API key needed.

What is AlphaGBM?

AlphaGBM is a real-data options & research intelligence layer for traders and AI agents. Every number comes from real market data -- IV, Greeks, VRP, skew, flow, plus a tracked research workspace -- not LLM hallucination.

These 26 skills bring AlphaGBM's capabilities into your AI workflow: Claude Code, Cursor, Windsurf, or any agent that supports skills.

Why AlphaGBM?

LLM Roleplay Tools Generic Finance APIs AlphaGBM
Data Source LLM-generated Delayed/basic Real-time options data
Verifiable "85% confidence" Partial Every number has a source
Options Depth None Basic chain IV/HV/VRP/Greeks/Skew/Surface
Scoring Subjective None Quantitative scoring (0-100 options, 1-10 stocks)
Analysis Model None None G = B + M (Gain = Basics + Momentum)
Battle-tested No Varies 10K users, 3mo live trading
Coverage US only Varies US + HK + CN + Commodities

Quick Start

Install as Claude Code Skills

# Clone into your project
git clone https://github.com/AlphaGBM/skills.git .claude/skills/alphagbm

# Or add as submodule
git submodule add https://github.com/AlphaGBM/skills.git .claude/skills/alphagbm

Install for Cursor

git clone https://github.com/AlphaGBM/skills.git .cursor/skills/alphagbm

Install CLI

# Clone and install
git clone https://github.com/AlphaGBM/skills.git
cd skills/cli
pip install -e .

# Set your API key
alphagbm config set-key agbm_xxxxxxxxxxxxxxxx

# Start analyzing
alphagbm stock analyze AAPL
alphagbm options score NVDA

See cli/README.md for full CLI documentation.

Try It (No API Key Needed)

All skills include built-in demo data for AAPL, NVDA, SPY, TSLA, and META. Just ask your AI:

"Analyze AAPL stock using AlphaGBM" "Score NVDA options" "Show me TSLA's volatility surface" "What's the best bullish strategy for META?"

Connect Live Data

# Set your API key for real-time data
export ALPHAGBM_API_KEY=agbm_xxxxxxxxxxxxxxxx
export ALPHAGBM_BASE_URL=https://alphagbm.zeabur.app  # optional, this is the default

# Get your free key at https://alphagbm.com/api-keys

Check API Health

curl https://alphagbm.zeabur.app/api/health

Returns API status, available data fields, data source health, and market coverage — no auth needed. Useful for AI agents to verify what's available before making calls.

Quota

Plan Stock Analysis Options Analysis Quick Quote / Snapshot
Free 2/day 1/day Unlimited
Plus 1,000/month 1,000/month Unlimited
Pro 5,000/month 5,000/month Unlimited

Skills Overview

Core Analysis (7 skills)

Skill What It Does Example Query
Stock Analysis G=B+M model: fundamentals, momentum, EV, risk score, AI report "Analyze AAPL"
Options Score Score 0-100 across 4 strategies (Sell Put/Call, Buy Put/Call) "Best NVDA call to buy"
Options Strategy Strategy builder + scanner with 15+ templates "Bullish play on TSLA"
Vol Surface 3D implied volatility across strikes & expiries "Is AAPL IV expensive?"
Vol Smile Skew analysis for a single expiration "NVDA put skew"
Greeks Greeks calculator + implied volatility solver "Greeks for AAPL 220C"
P&L Simulator What-if analysis for any position "Simulate my iron condor"

Data Intelligence (6 skills)

Skill What It Does Example Query
IV Rank IV percentile vs. 252-day history "Is TSLA IV high?"
Earnings IV Panel Crush history + implied move + IV Rank tag + priced Iron Condor "Iron Condor for META earnings"
Unusual Activity Smart money / large block detection "Unusual options flow today"
Market Sentiment VIX, Put/Call, Fear & Greed dashboard "Market sentiment now"
VIX Status 5-tier fear thermometer: calm / normal / seller sweet spot / caution / extreme fear "Is this a good time for BPS?"
FearScore Per-ticker 6-indicator panic composite; ≥60 is BPS entry signal "Fear score QQQ", "is NVDA oversold"

Workflow Tools (4 skills)

Skill What It Does Example Query
Compare Side-by-side stock & options comparison "AAPL vs MSFT"
Watchlist Monitor tickers for key changes "Add NVDA to watchlist"
Alert Set IV, price, or activity alerts "Alert if TSLA IV > 80"
Polymarket Prediction market vs. options pricing "Rate cut odds vs options"

Risk & Portfolio Discipline (4 skills) ✨

Exit, hedge, and sizing decisions quantified from real data — not opinion.

Skill What It Does Example Query
Hedge Advisor Scenario-driven hedge for an existing position (Falling Knife / Bottom Fishing / Gain Protection); returns priced Long Put / Collar / Tier-down specs "Hedge my AAPL at cost 140, now 180"
BPS Backtest Walk-forward backtest of Bull Put Spread with signal vs no-signal control in one call "Backtest BPS on QQQ — does FearScore work?"
Take-Profit Lab Any-ticker 15-strategy exit backtest; auto-classifies whether it's holdable or needs tiered exit via a novel "rollercoaster rate" metric "Should I hold TQQQ long-term?"
Duan-Yongping Analysis Three-panel seller playbook (Sell Put at willing-buy price / Covered Call yield / VIX-tier panic-buy context) "Duan-style analysis on AAPL"

Knowledge Base — Research Brain (5 skills)

Build a personal, monitored research workspace. Profiles auto-refresh, theses get checked against triggers, the system audits itself weekly.

Skill What It Does Example Query
Company Profile Auto-built research files: fundamentals, PE/PB band, red flags, event radar "Add NVDA to my knowledge base"
Investment Thesis Buy reasons + structured sell triggers, monitored automatically "Why did I buy AAPL?"
Macro View Track VIX / US10Y / DXY / gold with portfolio-aware impact analysis "Track VIX and US10Y"
Theme Research Group tickers into themes (AI infra, HK dividend) + news keyword watching "Create an AI infra theme"
Health Check Weekly audit: stale profiles, thesis drift, orphan pages → 0-100 score "Audit my research brain"

See Also

  • Investment Masters -- 10 masters' methodologies (Buffett, Dalio, Soros, Marks...) + 13F tracking

Architecture

You / Your AI Agent
    |  (natural language)
+------------------------------------------------------+
|              AlphaGBM Skills (this repo)              |
|                                                       |
|  Stock    Options   Vol      Strategy   Greeks   ...  |
|  Analysis  Score   Surface   Builder    Dashboard     |
+-------------------------+-----------------------------+
                          |
               +----------+----------+
               v                     v
         Mock Data              AlphaGBM API
      (built-in, free)      (alphagbm.zeabur.app)
                             Real-time market data
                             IV/HV/VRP/Greeks/Skew

How Skills Connect

Skills aren't isolated -- they reference each other to form a complete workflow:

Stock Analysis --> Options Score --> Options Strategy --> P&L Simulator
       |                |                    |
       v                v                    v
   Compare          Vol Surface           Greeks
                    Vol Smile
                    IV Rank --> Earnings Crush

Market Sentiment --> Unusual Activity --> Alert
                                          Watchlist

Polymarket --> Market Sentiment --> Options Strategy

Data Coverage

Market Stocks Options Data Points
US 200+ Full chains IV/HV/VRP/Greeks/Skew/Surface
HK 35+ Full chains IV/HV/VRP/Greeks
CN 20+ ETFs Full chains IV/HV/VRP/Greeks
Commodities Au/Ag/Cu/Al Futures options IV/Greeks/Delivery risk

Real Data, Not Guesswork

Every number in AlphaGBM is verifiable:

Metric Value How It's Computed
IV 32.5% Black-Scholes on actual bid/ask prices
IV Rank 58 Current IV vs. 252 trading days of history
VRP +4.0% Implied Vol - Historical Vol — measures option overpricing
Option Score 80/100 Weighted: premium yield + support/resistance + safety margin + trend + PoP + liquidity + time decay
Stock Score 7.0/10 G = B + M — Basics (PE, PEG, growth, margins) + Momentum (VIX, technicals, flow)
Risk 4/10 Additive: valuation +2, growth +2, liquidity +2, market +1.5, technical +1
EV +5.2% 50% × 1w + 30% × 1m + 20% × 3m expected value

This is not "based on my training data" or "I estimate with 85% confidence."

This is math on market data.

Example Workflow

You: "Analyze AAPL, then find the best options play"

The agent chains skills automatically:

1. GET  /api/stock/quick-quote/AAPL          → $261.40 (-0.8%)
2. POST /api/stock/analyze-sync              → G=B+M score 7.0/10, EV +5.2%, BUY
   {"ticker": "AAPL", "style": "balanced"}     Risk 4/10, target $275, stop-loss $239

3. GET  /api/options/snapshot/AAPL           → IV 32.5%, IV Rank 58, VRP +4.0%
4. POST /api/options/chain-sync              → Sell Put scores: 80, 78, 75...
   {"symbol": "AAPL", "expiry_date": "..."}    Buy Call scores: 76, 74, 72...

5. POST /api/options/tools/strategy/build    → Bull Call Spread 265/280
   {"template_id": "bull_call_spread"}         Max profit $1085, max loss $415

6. POST /api/options/tools/simulate          → Breakeven $269.15, PoP 44.5%
   {"symbol": "AAPL", "legs": [...]}

You: "Is that IV expensive?"

7. GET  /api/options/snapshot/AAPL           → IV Rank 58 (moderate)
8. GET  /api/options/tools/vol-surface/AAPL  → ATM IV in contango, earnings in 26d

All from real API calls. All verifiable.

Roadmap

  • 26 Skills with mock data
  • Claude Code & Cursor support
  • CLI tool (pip install -e ./cli)
  • Real-time WebSocket feeds
  • Community strategy sharing
  • More markets (EU, JP, KR options)

Contributing

See CONTRIBUTING.md for guidelines. We welcome:

  • Bug reports & feature requests
  • Skill improvements & new skill proposals
  • Translations (currently EN + CN)
  • Mock data for additional tickers

License

MIT -- see LICENSE.

Links


Built by the AlphaGBM team. Trusted by 10,000+ traders worldwide.

Real data. Real signals. Real edge.

About

Real-data options intelligence for AI agents — 15 Skills for Claude Code, Cursor & beyond

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 100.0%