Skip to main content

​
Using the AI Analyst

The AI Analyst is your personal options advisor - it combines mathematical risk metrics (Greeks) with live market data (sentiment, fundamentals, insider trading) to evaluate your strategy.
The AI Analyst uses LLMs (GPT-4 or OpenRouter models) to provide natural language analysis. It receives Position Greeks (total USD exposure), market context, and your strategy details to generate actionable insights.

​
How the AI Analyst Works

The system follows a three-stage analysis pipeline:
1

Calculate Mathematical Risk

The platform computes Position Greeks using Black-Scholes:
  • Delta: Directional exposure in USD
  • Gamma: Acceleration risk
  • Theta: Daily time decay in USD
  • Vega: Volatility sensitivity
These are already multiplied by 100 (contract size) and quantity.
2

Gather Market Context

The system fetches live market data:
  • Current spot price and analyst targets
  • VADER sentiment score from news/social media
  • Fundamental metrics (P/E, recommendations)
  • Insider trading flow (purchases vs. sales)
3

LLM Strategic Analysis

The AI receives:
  • Your strategy details (legs, strikes, expirations)
  • Position Greeks (not per-contract Greeks)
  • Market context data
  • Volume/Open Interest for liquidity assessment
It generates a 2-paragraph analysis evaluating:
  • Strategy fit vs. market sentiment
  • Risk metrics vs. conviction level
  • Liquidity concerns
  • Probability assessment

​
Accessing the AI Analyst

The AI Insights panel is on the right side of the interface with three tabs:

​
Tab 1: IA & Riesgo (AI & Risk)

This is where you get your AI analysis and risk metrics.
A danger meter that quantifies your position risk:Score Components:
  • Base score: 5 (neutral)
  • Heavy theta decay (< -20): +2 points
  • High directional leverage (|Δ| > 50): +1 point
  • Severe negative gamma (< -10): +3 points
  • Positive theta (> 20): -1 point
Interpretation:
  • 1-3: Low risk, conservative strategy
  • 4-6: Moderate risk, standard defined risk trades
  • 7-8: High risk, significant theta or gamma exposure
  • 9-10: Extreme risk, potential for rapid losses
Risk scores above 7 require smaller position sizes (1-2% of account max) and tight stop losses.
Always-visible grid showing net exposure:
Δ Delta: +16.09     Γ Gamma: -2.45
Θ Theta: +12.30/day V Vega: -8.75
Color coding:
  • Delta: Green if positive (bullish), red if negative (bearish)
  • Gamma: Red if negative (dangerous)
  • Theta: Green if positive (earning), red if negative (bleeding)
  • Vega: White (informational)
See Understanding Greeks for deep dive.
Automatic alerts based on Greek thresholds:Theta Alert:
“ALERTA: Riesgo de Theta acelerado. Pierdes $22.50/dia.”
→ Your negative theta is draining value quicklyDelta Alert:
“Cartera apalancada: Alcista (Delta +68.20).”
→ High bullish exposure, significant directional riskGamma Alert:
“ALERTA: Gamma negativo severo (Riesgo agudo de volatilidad).”
→ Rapid loss potential from price swingsPositive Theta:
“Ingresas $18.50/dia por primas desgastadas (Theta positiva).”
→ Earning daily income from time decay

​
Requesting AI Analysis

Click the ”🤖 Evaluacion Estrategica Global” button to trigger the LLM analysis.
1

Build Your Strategy First

The AI needs at least one option leg to analyze. Add your positions using the Strategy Builder.
Wait until your strategy is complete before requesting analysis. Making changes after will require re-running the AI.
2

Click the Analysis Button

The button shows ”⏳ Consultando Analista…” while processing (typically 5-15 seconds).The system:
  • Calculates fresh Greeks for current state
  • Bundles market context
  • Calls LiteLLM API with full context
  • Streams response back to UI
3

Read the Extended Analysis

A new panel appears with “Analisis Extendido IA” containing:
  • Strategy assessment vs. market sentiment
  • Risk evaluation against volatility conditions
  • Liquidity concerns (volume/OI analysis)
  • Probabilistic outlook
The analysis is limited to 2 concise paragraphs in fluent Spanish.
The AI analysis is based on current market data and mathematical projections. It is NOT financial advice. Always do your own research and never risk more than you can afford to lose.

​
Understanding AI Analysis Output

The LLM is prompted to analyze three critical dimensions:

​
1. Strategy vs. Sentiment Alignment

The AI crosses your directional bias (delta) against market sentiment: Example bullish strategy (Delta +45):
“La estrategia alcista con delta positivo de +45 se alinea bien con el sentimiento de mercado actual (VADER +0.18, Bullish). Los analistas mantienen un target de 520,un4520, un 4% por encima del spot actual de 500, lo que respalda la tesis direccional.”
Example bearish strategy with bullish sentiment:
“Cuidado: tu delta negativo (-38) apuesta a una caída, pero el sentimiento algorítmico es fuertemente alcista (+0.22). Los insiders han realizado 8 compras recientes vs. solo 2 ventas, señal contraria a tu posición.”

​
2. Risk Metrics vs. Conviction

The AI evaluates if your Greek exposure matches the strength of market signals: High conviction + high risk:
“El gamma negativo de -15 implica riesgo severo ante movimientos bruscos. Sin embargo, la baja volatilidad actual (VIX 16) y un rango consolidado en el precio justifican la exposición si mantienes stops ajustados en 495y495 y 510.”
Low conviction + high risk:
“ALERTA: Theta de -28 USD/día con un sentimiento neutral (VADER -0.02) es peligroso. Estás pagando alta prima sin una tesis direccional clara. Considera reducir tamaño o cambiar a spreads con theta positiva.”

​
3. Liquidity and Execution Concerns

The AI checks volume and open interest of your legs: Good liquidity:
“Liquidez excelente: el call $510 muestra volumen de 450 contratos y OI de 2,800. Entrada y salida serán fluidas con spreads ajustados.”
Poor liquidity:
“RIESGO DE LIQUIDEZ: el put $485 tiene apenas 12 contratos de volumen y 80 de OI. Spreads bid-ask amplios dificultarán el cierre de la posición. Considera strikes con mayor actividad.”

​
Tab 2: Contexto (Fundamental Data)

This tab displays fundamental metrics fetched from Yahoo Finance:
  • Symbol: Ticker identifier
  • Business Summary: Brief description of operations
  • Spot Price: Current market price
  • Recommendation: Analyst consensus (Buy, Hold, Sell)
Example Context: 
SPY Fundamentals
S&P 500 ETF Trust tracking the broad US equity market.

Spot: $500.25
Recommendation: Buy
Target Mean: $520.00
P/E Forward: 21.3
Interpretation: Analysts are bullish (target 4% higher), valuation is normal (P/E ~21). Supports moderately bullish strategies.

​
Tab 3: Sentimiento (Sentiment Analysis)

This tab shows algorithmic sentiment and insider trading data:
A compound score from -1 (extreme bearish) to +1 (extreme bullish) based on:
  • Recent news headlines (15-day window)
  • Social media mentions
  • Article tone analysis
Thresholds:
  • < -0.05: Bearish sentiment
  • -0.05 to +0.05: Neutral sentiment
  • > +0.05: Bullish sentiment
Visual: Circular badge with color coding:
  • Red: Bearish
  • Orange: Neutral
  • Green: Bullish
Up to 4 recent headlines used in sentiment calculation:
“Fed signals rate cuts may come sooner than expected"
"Tech earnings beat expectations, driving market rally"
"Inflation data comes in cooler, boosting investor confidence”
These provide context for the VADER score.
Corporate insider activity over recent months:Insider Purchases: Number of buying transactions
Insider Sales: Number of selling transactionsInterpretation:
  • More purchases than sales: Bullish signal (insiders confident)
  • More sales than purchases: Bearish signal (insiders cashing out)
  • Balanced: Neutral
Top Insiders: Details on largest transactions:
  • Name and position
  • Shares traded (+ for buy, - for sell)
Insider selling isn’t always bearish - executives often sell for tax/diversification. Look for unusual patterns (multiple insiders selling simultaneously).

​
Interpreting Combined Signals

The AI Analyst’s power comes from synthesizing all data sources:

​
Bullish Alignment Example

Your Strategy: Bull call spread (Delta +32)
Market Signals:
  • VADER Sentiment: +0.18 (Bullish)
  • Analyst Target: $520 (4% upside)
  • Insiders: 8 purchases, 2 sales
  • Volume: High liquidity
AI Output:
“Estrategia bien construida. El sesgo alcista (delta +32) coincide con el sentimiento positivo del mercado (+0.18) y el flujo de insiders (8 compras). El spread call limita riesgo mientras permite capturar el movimiento hacia el target de 520.Thetacontroladoen−8USD/dıˊaesmanejabledadoelhorizontede28dıˊas.Probabilidadfavorablesielpreciosupera520. Theta controlado en -8 USD/día es manejable dado el horizonte de 28 días. Probabilidad favorable si el precio supera 505 en 2 semanas.”

​
Bearish Conflict Example

Your Strategy: Iron condor (neutral, Delta +2)
Market Signals:
  • VADER Sentiment: +0.28 (Strong Bullish)
  • Analyst Target: $530 (6% upside)
  • Insiders: 12 purchases, 1 sale
  • Recent breakout from consolidation
AI Output:
“ALERTA: Tu estrategia neutral (iron condor) enfrenta riesgo de tendencia alcista fuerte. Sentimiento +0.28 y 12 compras de insiders sugieren momentum ascendente que puede testear tu call corto en 510.Aunquetienesthetapositivo(+18USD/dıˊa),elgammanegativo(−12)amplificaraˊpeˊrdidassielpreciorompe510. Aunque tienes theta positivo (+18 USD/día), el gamma negativo (-12) amplificará pérdidas si el precio rompe 505. Considera cerrar el lado call o ampliar strikes.”

​
Best Practices for AI Analysis

1

Build Complete Strategies

Add all legs before requesting analysis. Partial strategies generate incomplete insights.
2

Cross-Reference Tabs

Don’t just read the AI output. Check:
  • Fundamentals (Tab 2): Do analysts support your direction?
  • Sentiment (Tab 3): Is news flow aligned or conflicting?
  • Greeks (Tab 1): Are risk metrics acceptable?
3

Update After Major Events

Re-run analysis after:
  • Earnings releases
  • Fed announcements
  • Major news events
  • Significant price moves
Market context changes rapidly - stale analysis is dangerous.
4

Use as Confirmation, Not Conviction

The AI validates YOUR thesis. If the AI says “high risk” and you disagree, maybe you’re missing something. But if the AI says “looks good” and you’re uncertain, it’s not a green light to overlarge.

​
Common AI Warnings to Heed

“Gamma negativo severo”
Don’t ignore this. Negative gamma causes non-linear losses. Keep positions small and set stops.
“Riesgo de liquidez”
Illiquid options cost you money on both entry and exit. If the AI flags low volume/OI, choose different strikes.
“Sentimiento contrario a tu posición”
Fighting market sentiment is possible but requires strong conviction and tight risk management. The AI is telling you the odds are against you.

​
Limitations of AI Analysis

What the AI cannot predict:
  • Black swan events (sudden crashes, geopolitical shocks)
  • Earning surprises (better/worse than expected)
  • Fed policy pivots (surprise rate changes)
  • Market manipulation or unusual activity
  • Your personal risk tolerance and account size
Always apply human judgment and proper risk management regardless of AI output.

​
Advanced: Understanding the Prompt

The LLM receives a structured prompt: System Role:
“You are a Senior Options Analyst. The data shows Position Greeks (Total USD), already multiplied by 100 × quantity. A Delta of +16.09 means the entire position gains 16.09per16.09 per 1 stock move.”
Analysis Rules:
  1. Cross live market data (sentiment, analyst targets, insiders) with option liquidity (volume/OI)
  2. Explain if the strategy makes mathematical sense vs. current sentiment
  3. Limit to 2 impactful paragraphs in fluent Spanish
This ensures the AI:
  • Interprets Greeks correctly (total position, not per contract)
  • Focuses on actionable insights (not generic advice)
  • Maintains conciseness (no walls of text)

​
Pro Tips

  1. Run analysis multiple times as you adjust strikes/expirations to compare risk profiles
  2. Save AI output to a notes file for post-trade review (learn from predictions)
  3. Compare AI sentiment to your own research - if you see something the AI doesn’t, investigate deeper
  4. Pay attention to Quick Tips - these are hard-coded thresholds based on decades of options trading wisdom
  5. Use AI for education - the explanations teach you to spot these patterns yourself over time

​
Next Steps

The best traders use AI as a sanity check, not a decision-maker. Build your conviction through research, then validate it with AI analysis. If they align, you have a high-probability trade.

Build docs developers (and LLMs) love

Get started for freeTalk to us