10 AI Prompts That Actually Work for Stock & Crypto Analysis

Why AI Is Changing How Traders Do Research

Before large language models, retail traders had two options for stock research: spend hours reading 10-K filings, earnings transcripts, and analyst reports manually, or pay thousands per year for a Bloomberg terminal. Neither was practical for the average person managing their own portfolio.

AI trading prompts changed the equation. With ChatGPT, Claude, or Gemini, you can now paste a company's financial statements into a conversation and get a structured analysis in seconds. You can ask an LLM to stress-test your assumptions, identify risks you missed, or compare a stock against its sector peers -- tasks that used to take a junior analyst an entire afternoon.

But there is a catch. AI is only as useful as the prompt you give it. Ask "Is AAPL a good stock?" and you will get a wishy-washy summary that helps nobody. Ask it to "perform a three-scenario DCF analysis on Apple using the last four quarters of free cash flow, a WACC range of 8-12%, and terminal growth of 2-3%" and you get something genuinely valuable.

The prompts below are designed to extract maximum analytical value from any major LLM. They are structured, specific, and tested across hundreds of real analyses. Each one tells the AI exactly what role to play, what data to focus on, and what format to deliver the output in.

10 AI Trading Prompts You Can Copy and Use Today

These are organized into five categories that cover the full trading research workflow: fundamental analysis, risk assessment, market sentiment, portfolio review, and trade journaling. Each prompt is ready to paste directly into ChatGPT, Claude, or any other LLM.

Category 1: Fundamental Analysis Prompts

Fundamental analysis is where AI trading prompts deliver the biggest time savings. Instead of manually pulling ratios from financial statements, you give the AI raw data and let it do the heavy lifting.

Prompt #1 -- Deep Company Analysis

You are a senior equity research analyst. Analyze [COMPANY NAME]
(ticker: [TICKER]) as a potential investment. Cover these areas:

1. BUSINESS MODEL: Revenue streams, pricing model, customer
   segments, and geographic breakdown.
2. FINANCIAL HEALTH: Analyze the last 3 years of revenue growth,
   gross margin, operating margin, net margin, free cash flow,
   and debt-to-equity ratio. Flag any concerning trends.
3. COMPETITIVE MOAT: Score the company 1-5 on each: network
   effects, switching costs, cost advantages, intangible assets,
   efficient scale. Explain your scores.
4. VALUATION: Based on current P/E, EV/EBITDA, and P/FCF
   multiples, is the stock cheap, fair, or expensive relative
   to its sector peers?
5. RISKS: List the top 3 risks that could break the thesis.

Format: Use headers, bullet points, and a final BUY/HOLD/AVOID
recommendation with a one-sentence rationale.

Prompt #2 -- Earnings Call Breakdown

I am going to paste the transcript of [COMPANY]'s latest earnings
call. Analyze it and extract the following:

1. KEY METRICS: Revenue, EPS, and guidance vs. analyst consensus.
   Did they beat, meet, or miss on each?
2. MANAGEMENT TONE: Rate management's confidence on a scale of
   1-10 based on their language. Quote specific phrases that
   informed your rating.
3. FORWARD GUIDANCE: What did management guide for next quarter
   and full year? Any changes from prior guidance?
4. RED FLAGS: Any hedging language, unusual deferrals, executive
   departures mentioned, or accounting changes?
5. ANALYST QUESTIONS: Summarize the 3 most important analyst
   questions and whether management gave direct answers.

End with: "SIGNAL: [BULLISH/NEUTRAL/BEARISH]" and a 2-sentence
explanation.

[PASTE TRANSCRIPT BELOW]

Category 2: Risk Assessment Prompts

Most traders skip risk assessment because it is uncomfortable. AI trading prompts solve this by forcing structured risk evaluation without the emotional resistance. These ChatGPT trading prompts make you confront the downside before committing capital.

Prompt #3 -- Pre-Trade Risk Matrix

Act as a risk manager at a hedge fund. I am considering a
[LONG/SHORT] position in [TICKER] at $[PRICE].

Build a risk assessment matrix covering:

1. COMPANY-SPECIFIC RISKS: Management, concentration, balance
   sheet, litigation, insider activity.
2. INDUSTRY RISKS: Regulatory, disruption, cyclicality, new
   entrants.
3. MACRO RISKS: Interest rate sensitivity, recession impact,
   currency, geopolitical.

For each risk:
- Rate LIKELIHOOD: Low / Medium / High
- Rate IMPACT: Low / Medium / High
- Write one sentence: "If this materializes, [consequence]."

Then provide:
- Suggested position size (% of portfolio)
- Recommended stop-loss level and rationale
- Key catalyst dates to watch (earnings, FDA, etc.)

Flag any risk rated High/High as a potential deal-breaker.

Prompt #4 -- Bear Case Stress Test

I own [TICKER] and my thesis is: [DESCRIBE YOUR BULL CASE IN
2-3 SENTENCES].

Now play devil's advocate. You are a short-seller building the
strongest possible bear case against this company. Address:

1. Why is the current valuation unjustified?
2. What structural problems does management not talk about?
3. Which competitors are gaining ground and why?
4. What macro scenario would crush this stock specifically?
5. What is the realistic downside price target if the bear
   case plays out, and what assumptions drive that number?

Be ruthless. Do not hedge. I need the strongest possible
counterargument to my thesis so I can stress-test my conviction.

Category 3: Market Sentiment Prompts

Sentiment analysis is one of the highest-value applications of AI prompts for stock analysis. LLMs can process large volumes of text -- news, social media posts, analyst commentary -- and extract the prevailing mood faster than any human could.

Prompt #5 -- Sector Sentiment Scanner

Analyze the current market sentiment for the [SECTOR NAME]
sector (e.g., semiconductors, biotech, clean energy). Cover:

1. INSTITUTIONAL POSITIONING: Based on recent 13F filings and
   fund flow data you're aware of, are institutions adding or
   reducing exposure to this sector?
2. NARRATIVE ANALYSIS: What are the top 3 bullish narratives
   and top 3 bearish narratives currently driving this sector?
3. VALUATION CONTEXT: Is the sector trading above or below its
   5-year average P/E and EV/EBITDA? What's priced in?
4. CATALYST CALENDAR: List upcoming events in the next 90 days
   that could shift sentiment (earnings clusters, regulatory
   decisions, macro data releases).
5. CONTRARIAN SIGNAL: If sentiment is overwhelmingly bullish,
   what's the contrarian bear case? If bearish, what's the
   contrarian bull case?

Output a SENTIMENT SCORE from -5 (extreme fear) to +5 (extreme
greed) and explain what a score at this level has historically
meant for forward returns.

Prompt #6 -- News Impact Assessment

A major news event just happened: [DESCRIBE THE NEWS EVENT].

Assess the impact on [TICKER / SECTOR]:

1. IMMEDIATE IMPACT: How will this affect the stock price in
   the next 1-5 trading days? Is the move likely priced in
   already?
2. SECOND-ORDER EFFECTS: What downstream consequences might
   the market be underpricing? (Supply chain, competitors,
   regulatory response.)
3. HISTORICAL PARALLELS: Name 2-3 similar past events and
   what happened to the stock/sector over the following 30,
   90, and 180 days.
4. TRADE SETUP: Based on this analysis, what is the
   highest-probability trade? Entry, target, stop-loss.

Clearly state your confidence level: HIGH / MEDIUM / LOW.

Category 4: Portfolio Review Prompts

Reviewing your own portfolio objectively is nearly impossible. Endowment effect, anchoring bias, and loss aversion cloud every self-assessment. AI trading prompts let you hand your portfolio to an unbiased analyst and get honest feedback.

Prompt #7 -- Portfolio Audit

You are a portfolio manager conducting a quarterly review.
Here is my current portfolio:

[LIST EACH POSITION: Ticker, # shares, avg cost, current price,
% of portfolio]

Analyze:

1. CONCENTRATION RISK: Am I overweight in any single stock,
   sector, or geography? Flag anything above 20% in one name
   or 40% in one sector.
2. CORRELATION ANALYSIS: Which of my positions are likely to
   move together? Am I diversified or do I just own 10 versions
   of the same trade?
3. RISK/REWARD RANKING: Rank each position from best to worst
   risk/reward at CURRENT prices (not my cost basis -- that's
   a sunk cost).
4. REBALANCING SUGGESTIONS: If I had to cut 2 positions and
   add 1, which would you cut and what would you add, given
   the current macro environment?
5. CASH ALLOCATION: Am I holding too much or too little cash
   given current market conditions?

Be direct. Do not sugar-coat underperforming positions.

Prompt #8 -- Crypto Portfolio Rebalance

Review my crypto portfolio and suggest rebalancing:

[LIST EACH POSITION: Token, amount, avg entry price, current
price, % of crypto allocation]

My risk tolerance is [CONSERVATIVE / MODERATE / AGGRESSIVE].
My time horizon is [SHORT: <6mo / MEDIUM: 6-18mo / LONG: 18mo+].

Analyze:
1. Layer 1 vs Layer 2 vs DeFi vs meme coin exposure breakdown.
2. Stablecoin allocation -- is it appropriate for current
   market conditions?
3. Which positions have the weakest fundamentals relative to
   their portfolio weight?
4. Suggest a target allocation with percentages and reasoning.
5. Identify any tokens I should consider adding for
   diversification.

Factor in the current Bitcoin dominance level and where we are
in the market cycle.

Category 5: Trade Journaling Prompts

Trade journaling is the single most underused tool among retail traders. Most people skip it because it is tedious. AI prompts for stock analysis can turn a two-sentence trade note into a structured journal entry that actually helps you improve.

Prompt #9 -- Post-Trade Analysis

I just closed a trade. Help me analyze it for my trading
journal.

TRADE DETAILS:
- Ticker: [TICKER]
- Direction: [LONG / SHORT]
- Entry: $[PRICE] on [DATE]
- Exit: $[PRICE] on [DATE]
- Position size: [% OF PORTFOLIO]
- P&L: [AMOUNT or %]

My original thesis was: [2-3 SENTENCES]

Analyze:
1. THESIS ACCURACY: Was my original thesis correct, partially
   correct, or wrong? What actually drove the price move?
2. TIMING: Did I enter too early, too late, or at the right
   time? Same for the exit.
3. POSITION SIZING: Was the size appropriate given the risk?
4. EMOTIONAL CHECK: Based on my entry/exit timing, are there
   signs of FOMO, panic, or overconfidence?
5. LESSON: State the single most important takeaway from this
   trade in one sentence.
6. PATTERN: Does this trade resemble any common trading mistake
   (revenge trading, averaging down into a loser, cutting
   winners too early, etc.)?

Rate the overall trade execution: A / B / C / D / F.

Prompt #10 -- Weekly Trading Review

Conduct a weekly review of my trading activity.

This week's trades:
[LIST: Date, Ticker, Direction, Entry, Exit, P&L]

Open positions entering next week:
[LIST: Ticker, Direction, Entry, Current Price, Unrealized P&L]

Analyze:
1. WIN RATE: What percentage of trades were profitable?
2. RISK/REWARD RATIO: Average winner size vs average loser
   size. Am I cutting losers fast enough?
3. BEST TRADE: Which trade had the best execution and why?
4. WORST TRADE: Which trade had the worst execution and what
   should I have done differently?
5. PATTERN RECOGNITION: Are there recurring mistakes across
   multiple trades this week?
6. PLAN FOR NEXT WEEK: Based on open positions and current
   market conditions, what should I focus on?

Keep the tone constructive but honest. I want to improve, not
feel good about bad trades.

How to Structure AI Trading Prompts for Better Results

The 10 prompts above follow a specific framework that you can apply when writing your own ChatGPT trading prompts. Whether you are analyzing stocks, crypto, options, or commodities, these structural principles will dramatically improve the quality of AI output.

1. Assign a Role

Starting a prompt with "You are a senior equity research analyst" or "Act as a risk manager at a hedge fund" is not just theater. It primes the model to use domain-specific vocabulary, frameworks, and analytical depth. A prompt without a role assignment produces consumer-grade answers. A prompt with one produces analyst-grade output.

2. Provide Specific Data

The single biggest mistake in AI prompts for stock analysis is asking the model to "look up" financial data. LLMs have knowledge cutoffs and can hallucinate numbers. Instead, paste the actual data -- financial statements, earnings transcripts, trade logs, portfolio holdings. The model analyzes what you give it; do not rely on what it "knows."

3. Define the Output Format

Tell the AI exactly how you want the response structured: numbered lists, tables, bullet points, a final rating, a specific score range. Without format instructions, you get walls of text. With them, you get scannable, actionable output you can reference quickly during trading hours.

4. Include Constraints

Constraints improve focus. "List the top 3 risks" is better than "list the risks" because it forces prioritization. "Rate confidence as HIGH / MEDIUM / LOW" forces the model to commit to a conviction level instead of hedging everything. "Be ruthless. Do not hedge." removes the model's default tendency toward diplomatic non-answers.

5. Chain Prompts for Depth

A single prompt can only do so much. The best AI trading workflows chain multiple prompts in sequence: start with a broad company analysis, then drill into the DCF, then run the bear case stress test, then do a sentiment check. Each subsequent prompt builds on the context from the previous one. This mirrors how professional analysts actually work -- layer by layer, not all at once.

Combining AI Analysis with Real-Time Technical Data

AI trading prompts excel at qualitative analysis -- interpreting earnings calls, evaluating competitive moats, stress-testing theses. But they cannot give you live market data. RSI, MACD, Bollinger Bands, and price action need to come from a real data source.

This is where combining AI prompts with a technical analysis API creates a workflow that is greater than the sum of its parts. The MarginLab provides free real-time indicator data that you can feed directly into your AI prompts for a complete analysis:

Step 1: Get technical data from the API
────────────────────────────────────────
GET /api/analyze?symbol=AAPL&days=90

Response:
{
  "analysis": {
    "rsi": 62.4,
    "macd": { "macd": 3.21, "signal": 2.88 },
    "bollinger": { "upper": 242, "middle": 228, "lower": 214 },
    "overall": "BULLISH",
    "score": 3
  }
}

Step 2: Feed it into your AI prompt
────────────────────────────────────
"Here is the current technical picture for AAPL:
RSI: 62.4, MACD: bullish crossover (3.21 vs 2.88 signal),
Bollinger: price near middle band at $228.
Combined with the fundamental analysis above, what is
the optimal entry strategy and position sizing?"

This hybrid approach -- quantitative data from an API plus qualitative analysis from AI prompts -- is how the best independent traders operate in 2026. The API handles the math that LLMs are unreliable at (precise calculations, real-time prices), while the LLM handles the pattern recognition and synthesis that APIs cannot do (interpreting management tone, evaluating competitive dynamics, generating trade ideas).

For a deeper dive into the technical side, read our guide on AI-powered crypto technical analysis or explore the free trading calculators for position sizing and risk management.

What NOT to Do with AI Trading Prompts

AI is a powerful research tool, but it has clear limitations. These are the mistakes that will cost you money if you are not careful:

• Do not treat AI output as financial advice. An LLM is not a licensed financial advisor. It does not know your tax situation, risk tolerance, time horizon, or full financial picture. Use it as one input among many, not as the decision-maker.
• Do not trust AI-generated numbers without verification. LLMs can and do hallucinate financial data. If a prompt response includes specific revenue figures, earnings numbers, or historical prices, verify them against an authoritative source (SEC filings, Yahoo Finance, the company's investor relations page). Never trade on unverified numbers.
• Do not ask for price predictions. "Where will TSLA be in 6 months?" is a useless prompt. No model -- AI or otherwise -- can predict future prices. What AI can do is help you evaluate the probability-weighted range of outcomes and position accordingly.
• Do not skip your own due diligence. AI should accelerate your research, not replace it. If an AI analysis says "buy" but you have not read the 10-K yourself, you do not understand the position well enough to hold it through volatility. AI generates the first draft; you do the final review.
• Do not use a single prompt in isolation. One prompt gives you one perspective. Layer multiple prompts -- bullish analysis, then bear case stress test, then risk assessment -- to build a three-dimensional view. The prompts in this guide are designed to work together as a complete research workflow.
• Do not ignore the knowledge cutoff. Every LLM has a training data cutoff. It may not know about the most recent earnings report, a CEO departure last week, or a regulatory ruling from yesterday. Always supplement AI analysis with current news and real-time data from sources like the MarginLab.

Go Beyond These 10: The Full 100 AI Trading Prompts Pack

The prompts in this guide are a starting point. They cover the core workflows, but experienced traders need more specialized tools: options analysis prompts, sector rotation signals, earnings season workflows, crypto tokenomics evaluation, IPO analysis frameworks, and macro regime identification.

The AI Trading Prompts Pack contains 100 tested prompts organized into 10 categories, each one refined through real trading research. It includes everything in this guide plus:

• Advanced DCF and valuation prompts with scenario modeling
• Options strategy analysis and Greeks interpretation
• Crypto tokenomics and on-chain analysis prompts
• Macro regime identification and sector rotation signals
• Earnings season playbook (pre-earnings, reaction, post-earnings)
• IPO and SPAC analysis frameworks
• Automated weekly and monthly portfolio review templates
• Prompt chaining sequences for end-to-end research workflows

If you want to pair AI analysis with technical signals and fundamental research tools, the Complete Trading Bundle includes the AI Prompts Pack, the Stock Analysis Toolkit, and every other MarginLab product at a significant discount.