STRATOSFI

Appearance

AI Tips & Best Practices ​

Stratosfi's AI assistant can analyze stocks, explain concepts, compare companies, help build dashboards, and explore company relationships. The quality of the answers depends heavily on how you ask. This guide shows you how to get the most out of every interaction.

Writing Better Prompts ​

The AI performs best when you give it specific, focused questions with enough context to understand what you need.

Be Specific ​

Vague prompts get generic answers. Specific prompts get actionable analysis.

Instead of...Try...
"Tell me about Apple""Analyze AAPL's valuation — is it expensive relative to its 5-year average P/E?"
"What should I buy?""Compare MSFT, GOOGL, and AMZN on revenue growth, margins, and forward P/E"
"How's the market?""What sectors are outperforming the S&P 500 this month and why?"
"Is this a good stock?""What are the bull and bear cases for NVDA at its current price?"

Include the Ticker ​

Always include the ticker symbol when asking about a specific stock. The AI uses the ticker to pull real-time data and give you a grounded answer.

  • "Analyze TSLA" pulls live price, fundamentals, and recent news for Tesla.
  • "Analyze Tesla" also works, but the ticker removes any ambiguity.

Ask for Comparisons ​

Some of the most useful AI outputs come from comparative analysis:

  • "Compare AMD vs INTC vs NVDA — which has the best valuation for its growth rate?"
  • "How does JPM's dividend yield compare to other big banks?"
  • "Rank the FAANG stocks by free cash flow yield."

Ask Follow-Up Questions ​

The AI remembers context within a conversation. Use this to drill deeper:

  1. "Analyze AAPL" — get the overview.
  2. "What are the biggest risks to that thesis?" — dig into the bear case.
  3. "How does its supply chain in China affect those risks?" — go deeper.
  4. "What would be a good entry point based on the chart?" — move to action.

What to Ask ​

Stock Analysis ​

  • "Give me a bull and bear case for AMZN"
  • "Is META overvalued or undervalued based on its growth rate?"
  • "What is COST's competitive moat?"
  • "Summarize NFLX's last earnings report — did they beat expectations?"
  • "What are the key risks for BA right now?"

Portfolio Guidance ​

  • "My portfolio is 60% tech, 20% healthcare, 20% cash. Is that balanced?"
  • "I own AAPL, MSFT, GOOGL, AMZN, and NVDA. How correlated are these positions?"
  • "Suggest sectors I should add for better diversification."
  • "I am up 40% on TSLA. Should I take some profits or let it ride?"

Macro & Economy ​

  • "How does the current interest rate environment affect growth stocks?"
  • "What happens to utility stocks when the Fed cuts rates?"
  • "Explain the relationship between the yield curve and recessions."
  • "Which emerging markets look attractive right now and why?"

Learning & Concepts ​

  • "Explain the RSI indicator like I am a beginner"
  • "What is the difference between P/E and forward P/E?"
  • "How do options Greeks interact with each other?"
  • "What is a golden cross and how reliable is it as a signal?"
  • "Explain dollar-cost averaging with a real example"

Crypto-Specific ​

  • "Analyze BTC's price action this week"
  • "What is driving SOL's outperformance against ETH?"
  • "Explain DeFi yield farming in simple terms"
  • "What are funding rates and why do they matter?"

What NOT to Ask ​

The AI is a research and analysis tool, not a crystal ball. Understanding its limitations helps you use it properly.

Do Not Ask for Predictions ​

  • "Will AAPL go up tomorrow?" — Nobody knows. The AI will not pretend to.
  • "What will BTC be worth in 2027?" — Price targets far in the future are speculation, not analysis.
  • "Tell me a stock that will 10x" — If anyone could reliably do this, they would not be writing software.

Do Not Ask for Financial Advice ​

  • "Should I put my savings into TSLA?" — The AI cannot know your financial situation, risk tolerance, or goals.
  • "Is it a good time to buy a house?" — This is personal financial advice, not market analysis.

The AI provides analysis and information. The investment decisions are yours.

Do Not Expect Proprietary Data ​

The AI works with publicly available market data. It does not have access to:

  • Insider trading data or non-public filings
  • Real-time order flow or Level 2 data
  • Proprietary research from investment banks
  • Your broker account details

Do Not Take AI Output as Fact Without Verification ​

The AI can make mistakes, especially with very recent events or obscure data points. Use it as a starting point for your own research, not as the final word.

Using AI to Build Dashboards ​

The AI Dashboard Builder in Custom Dashboards creates layouts from natural language descriptions. Here is how to get the best results.

Describe the Purpose, Not Just the Widgets ​

Weak promptStrong prompt
"Add a chart and a watchlist""Build a dashboard for monitoring oil and gas stocks — I want to see crude oil prices, XLE sector performance, and charts for my top 5 energy holdings"
"Make a crypto dashboard""Create a crypto trading dashboard with BTC and ETH charts, funding rates, top movers by 24h volume, and a news feed filtered to crypto"

Include Specific Tickers When Possible ​

"Build a FAANG monitoring dashboard with charts for AAPL, AMZN, GOOGL, META, and NFLX, plus a combined news feed and a peer comparison table."

The AI can set up each widget pre-configured to the right tickers.

Iterate ​

After the AI generates a dashboard:

  1. Review what it created.
  2. Ask it to adjust: "Add a correlation matrix for these stocks" or "Replace the watchlist with a heatmap."
  3. Fine-tune widget positions and sizes by dragging.

Example Prompts for Dashboards ​

  • "Build a morning market overview dashboard — show S&P 500, Nasdaq, and Dow charts, today's top movers, economic calendar for the week, and a news feed"
  • "Create a dashboard for tracking my earnings plays — I need charts for AAPL, MSFT, GOOGL, AMZN with earnings dates highlighted, plus an earnings calendar widget"
  • "Set up a macro analysis dashboard — yield curves, sector rotation heatmap, central bank rates, and an economic indicators panel"
  • "Build a risk monitoring dashboard for my portfolio — sector exposure pie chart, correlation matrix, portfolio beta, and VaR metrics"

Using AI in Company Explorer ​

The Company Explorer includes an AI-powered sidebar that enhances graph exploration with natural language.

Expanding the Graph ​

Instead of manually clicking through nodes, tell the AI what you want to see:

  • "Show me Apple's major suppliers"
  • "Who are Tesla's competitors in the EV space?"
  • "Expand the supply chain for NVDA — focus on semiconductor equipment suppliers"
  • "Show me the private equity firms connected to this sector"

Analyzing Relationships ​

Ask the AI to interpret the graph for you:

  • "What is the risk if TSMC has production issues?" — The AI traces the dependency chain.
  • "Which companies in this graph have the highest correlation with AAPL?"
  • "Are there any single points of failure in this supply chain?"
  • "Which of these competitors is best positioned for growth?"

Building an Investment Thesis ​

Use the explorer and AI together to build a narrative:

  1. Start with a company: "Explore NVDA"
  2. Ask about the ecosystem: "Who are NVDA's biggest customers by revenue?"
  3. Follow the chain: "Now show me AMD's customers for comparison"
  4. Synthesize: "Based on this graph, which company has more diversified revenue sources?"

Example Prompts by Category ​

Quick Analysis (under 30 seconds) ​

  • "What is AAPL trading at and how has it moved this week?"
  • "Give me a one-paragraph summary of TSLA"
  • "What is the RSI on NVDA right now?"
  • "Top 3 gainers in the S&P 500 today"

Deep Analysis (1-2 minutes) ​

  • "Do a full fundamental analysis of MSFT — valuation, growth, margins, balance sheet, and competitive position"
  • "Compare the semiconductor sector vs. software sector on valuation and growth metrics"
  • "Analyze AMZN's revenue segments — which is growing fastest and which has the best margins?"
  • "Review my portfolio (AAPL, GOOGL, NVDA, JPM, JNJ, XOM) for concentration risk and suggest improvements"

Strategy Exploration ​

  • "What stock screening criteria would Warren Buffett likely use?"
  • "Design a dividend growth strategy — what should I screen for?"
  • "How would I build a barbell portfolio with growth and value stocks?"
  • "What is a pairs trading strategy and how would I find good pairs using the correlation matrix?"

Learning Sequences ​

Start broad, then go deep:

  1. "What are the most important fundamental metrics for evaluating a stock?"
  2. "Explain P/E ratio in depth — what is a good P/E and when is it misleading?"
  3. "How do I compare P/E ratios across different sectors?"
  4. "Show me how AAPL's P/E compares to the tech sector average"

Limitations and Disclaimers ​

What the AI Can Do ​

  • Analyze publicly available market data in real time
  • Explain financial concepts clearly
  • Compare companies on objective metrics
  • Identify patterns and summarize trends
  • Help you build dashboards and explore company relationships
  • Process follow-up questions within a conversation

What the AI Cannot Do ​

  • Predict future prices or market movements
  • Access non-public information
  • Replace a licensed financial advisor
  • Guarantee any investment outcome
  • Account for your personal financial situation

Token Usage ​

AI interactions consume tokens based on your subscription plan. Longer conversations and complex analyses use more tokens. To conserve tokens:

  • Ask focused questions instead of open-ended ones.
  • Use follow-ups within the same conversation instead of starting new ones.
  • For simple data lookups (current price, basic stats), use the Quote Panel or Screener instead of the AI.

The Bottom Line ​

The AI is a powerful research assistant, but it is not infallible and it is not a financial advisor. Use it to accelerate your analysis, learn faster, and explore ideas — then make your own informed decisions.

STRATOSFI — Smart Trading Platform

Built for traders, by traders