Build a Stock Analysis Agent with AutoGPT (Fundamentals + News)

pip install yfinance autogen-agentchat openai newspaper3k requests beautifulsoup4 pandas

stock_agent/
├── main.py
├── tools/
│   ├── __init__.py
│   ├── yahoo_finance.py
│   └── news_fetcher.py
├── prompts/
│   └── analyst_system.txt
└── output/

# tools/yahoo_finance.py
import yfinance as yf
import pandas as pd
from typing import Dict, Any, Optional
import json


def get_stock_fundamentals(ticker: str) -> Dict[str, Any]:
    """Fetch key fundamental data for a stock ticker."""
    try:
        stock = yf.Ticker(ticker)
        info = stock.info

        # Extract key metrics with safe fallbacks
        fundamentals = {
            "ticker": ticker.upper(),
            "company_name": info.get("longName", "Unknown"),
            "sector": info.get("sector", "Unknown"),
            "industry": info.get("industry", "Unknown"),
            "market_cap": info.get("marketCap"),
            "current_price": info.get("currentPrice") or info.get("regularMarketPrice"),
            "52_week_high": info.get("fiftyTwoWeekHigh"),
            "52_week_low": info.get("fiftyTwoWeekLow"),
            "pe_ratio": info.get("trailingPE"),
            "forward_pe": info.get("forwardPE"),
            "pb_ratio": info.get("priceToBook"),
            "ps_ratio": info.get("priceToSalesTrailing12Months"),
            "ev_to_ebitda": info.get("enterpriseToEbitda"),
            "profit_margin": info.get("profitMargins"),
            "revenue_growth": info.get("revenueGrowth"),
            "earnings_growth": info.get("earningsGrowth"),
            "debt_to_equity": info.get("debtToEquity"),
            "current_ratio": info.get("currentRatio"),
            "return_on_equity": info.get("returnOnEquity"),
            "return_on_assets": info.get("returnOnAssets"),
            "dividend_yield": info.get("dividendYield"),
            "analyst_target_price": info.get("targetMeanPrice"),
            "recommendation": info.get("recommendationMean"),
            "number_of_analysts": info.get("numberOfAnalystOpinions"),
            "business_summary": info.get("longBusinessSummary", "")[:500],
        }

        return {"status": "success", "data": fundamentals}

    except Exception as e:
        return {"status": "error", "message": str(e)}


def get_price_history(ticker: str, period: str = "1y") -> Dict[str, Any]:
    """Get historical price data and calculate basic metrics."""
    try:
        stock = yf.Ticker(ticker)
        hist = stock.history(period=period)

        if hist.empty:
            return {"status": "error", "message": "No price data found"}

        # Calculate returns
        start_price = float(hist["Close"].iloc[0])
        end_price = float(hist["Close"].iloc[-1])
        total_return = ((end_price - start_price) / start_price) * 100

        # Volatility (annualized)
        daily_returns = hist["Close"].pct_change().dropna()
        annualized_volatility = float(daily_returns.std() * (252 ** 0.5) * 100)

        # Average volume
        avg_volume = float(hist["Volume"].mean())

        summary = {
            "period": period,
            "start_price": round(start_price, 2),
            "end_price": round(end_price, 2),
            "total_return_pct": round(total_return, 2),
            "annualized_volatility_pct": round(annualized_volatility, 2),
            "average_daily_volume": int(avg_volume),
            "data_points": len(hist),
        }

        return {"status": "success", "data": summary}

    except Exception as e:
        return {"status": "error", "message": str(e)}


def get_financial_statements(ticker: str) -> Dict[str, Any]:
    """Get simplified income statement and balance sheet data."""
    try:
        stock = yf.Ticker(ticker)

        # Income statement (annual)
        income_stmt = stock.income_stmt
        balance_sheet = stock.balance_sheet

        result = {}

        if not income_stmt.empty:
            latest_year = income_stmt.columns[0]
            result["income_latest_year"] = str(latest_year.date())
            result["total_revenue"] = _safe_float(income_stmt, "Total Revenue", latest_year)
            result["gross_profit"] = _safe_float(income_stmt, "Gross Profit", latest_year)
            result["operating_income"] = _safe_float(income_stmt, "Operating Income", latest_year)
            result["net_income"] = _safe_float(income_stmt, "Net Income", latest_year)

        if not balance_sheet.empty:
            latest_quarter = balance_sheet.columns[0]
            result["total_assets"] = _safe_float(balance_sheet, "Total Assets", latest_quarter)
            result["total_debt"] = _safe_float(balance_sheet, "Total Debt", latest_quarter)
            result["cash"] = _safe_float(balance_sheet, "Cash And Cash Equivalents", latest_quarter)

        return {"status": "success", "data": result}

    except Exception as e:
        return {"status": "error", "message": str(e)}


def _safe_float(df, row_name: str, col) -> Optional[float]:
    try:
        val = df.loc[row_name, col]
        return float(val) if pd.notna(val) else None
    except (KeyError, TypeError):
        return None

# tools/news_fetcher.py
import yfinance as yf
from typing import List, Dict


def get_stock_news(ticker: str, max_articles: int = 10) -> Dict:
    """Fetch recent news headlines for a stock ticker."""
    try:
        stock = yf.Ticker(ticker)
        news = stock.news

        articles = []
        for item in (news or [])[:max_articles]:
            articles.append({
                "title": item.get("title", ""),
                "publisher": item.get("publisher", ""),
                "link": item.get("link", ""),
                "published": item.get("providerPublishTime", 0),
                "type": item.get("type", ""),
                "summary": item.get("summary", "")[:300] if item.get("summary") else "",
            })

        return {
            "status": "success",
            "ticker": ticker.upper(),
            "article_count": len(articles),
            "articles": articles,
        }

    except Exception as e:
        return {"status": "error", "message": str(e)}

# main.py
import autogen
import json
import os
from datetime import datetime
from tools.yahoo_finance import get_stock_fundamentals, get_price_history, get_financial_statements
from tools.news_fetcher import get_stock_news


def analyze_stock(ticker: str, output_dir: str = "output") -> str:
    os.makedirs(output_dir, exist_ok=True)

    llm_config = {
        "config_list": [
            {
                "model": "gpt-4o",
                "api_key": os.getenv("OPENAI_API_KEY"),
            }
        ],
        "temperature": 0.1,
        "functions": [
            {
                "name": "get_fundamentals",
                "description": "Fetch fundamental financial data for a stock ticker",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "ticker": {"type": "string", "description": "Stock ticker symbol"}
                    },
                    "required": ["ticker"]
                }
            },
            {
                "name": "get_price_history",
                "description": "Get 1-year price history and return metrics",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "ticker": {"type": "string"}
                    },
                    "required": ["ticker"]
                }
            },
            {
                "name": "get_news",
                "description": "Get recent news articles about the stock",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "ticker": {"type": "string"},
                        "max_articles": {"type": "integer", "default": 8}
                    },
                    "required": ["ticker"]
                }
            },
            {
                "name": "get_financials",
                "description": "Get income statement and balance sheet data",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "ticker": {"type": "string"}
                    },
                    "required": ["ticker"]
                }
            },
        ]
    }

    analyst = autogen.AssistantAgent(
        name="StockAnalyst",
        system_message="""You are a financial research analyst. Your job is to produce
structured research reports based on publicly available financial data.

IMPORTANT DISCLAIMER: You must always include a prominent disclaimer that your analysis
is for informational purposes only and does not constitute investment advice.

When analyzing a stock, follow this process:
1. Fetch fundamental data using get_fundamentals
2. Get 1-year price history using get_price_history
3. Get financial statements using get_financials
4. Fetch recent news using get_news
5. Synthesize all data into a structured markdown report

Your report must include:
- Company overview
- Key metrics table (P/E, P/B, debt ratios, margins)
- 1-year price performance summary
- Recent news summary (3-5 key themes)
- Basic valuation context (compare P/E to sector average if known)
- Risks and considerations
- Disclaimer

End your response with ANALYSIS_COMPLETE when done.""",
        llm_config=llm_config,
    )

    # Function map for the user proxy
    function_map = {
        "get_fundamentals": lambda ticker: get_stock_fundamentals(ticker),
        "get_price_history": lambda ticker: get_price_history(ticker),
        "get_news": lambda ticker, max_articles=8: get_stock_news(ticker, max_articles),
        "get_financials": lambda ticker: get_financial_statements(ticker),
    }

    user_proxy = autogen.UserProxyAgent(
        name="ResearchCoordinator",
        human_input_mode="NEVER",
        max_consecutive_auto_reply=15,
        is_termination_msg=lambda msg: "ANALYSIS_COMPLETE" in (msg.get("content") or ""),
        function_map=function_map,
        code_execution_config=False,
    )

    user_proxy.initiate_chat(
        analyst,
        message=f"Please conduct a comprehensive research analysis of {ticker.upper()} stock. Use all available tools to gather data and produce a structured report.",
    )

    # Extract the final report
    messages = analyst.chat_messages.get(user_proxy, [])
    final_report = ""
    for msg in reversed(messages):
        if msg.get("role") == "assistant" and msg.get("content"):
            final_report = msg["content"].replace("ANALYSIS_COMPLETE", "").strip()
            break

    # Save the report
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    output_file = os.path.join(output_dir, f"{ticker.upper()}_analysis_{timestamp}.md")

    with open(output_file, "w", encoding="utf-8") as f:
        f.write(f"# Stock Analysis: {ticker.upper()}\n")
        f.write(f"*Generated: {datetime.now().strftime('%Y-%m-%d %H:%M UTC')}*\n\n")
        f.write("---\n\n")
        f.write(final_report)

    print(f"Analysis saved to: {output_file}")
    return output_file


if __name__ == "__main__":
    import sys
    ticker = sys.argv[1] if len(sys.argv) > 1 else "AAPL"
    analyze_stock(ticker)

# Stock Analysis: AAPL
*Generated: 2026-05-31 14:30 UTC*

---

## Company Overview
Apple Inc. (AAPL) is a consumer electronics and software company...

## Key Metrics

| Metric | Value | Context |
|--------|-------|---------|
| P/E Ratio (TTM) | 28.4 | Above tech sector avg (~25) |
| Forward P/E | 25.1 | More reasonable on growth basis |
| P/B Ratio | 42.1 | Premium brand valuation |
| Debt/Equity | 1.73 | Moderate leverage |
| Profit Margin | 25.3% | Strong operating efficiency |
| Revenue Growth (YoY) | 4.2% | Moderate growth |
| Analyst Target | $198.50 | Based on 42 analyst estimates |

## 1-Year Price Performance
- Start: $165.23 → Current: $182.41
- Total Return: +10.4%
- Annualized Volatility: 22.1%

## Recent News Themes
1. **AI Integration:** Multiple reports on Apple Intelligence rollout...
2. **Services Growth:** App Store revenue up in Q1 2026...

## ⚠️ DISCLAIMER
This report is generated by an AI system for informational purposes only.
It does not constitute investment advice...

export OPENAI_API_KEY=sk-...
python main.py MSFT

# Create a watchlist file
echo "AAPL\nMSFT\nGOOGL\nAMZN" > watchlist.txt

# Run analysis on each
while read ticker; do
  python main.py $ticker
  sleep 10  # Rate limit courtesy delay
done < watchlist.txt