stock market consulting ai project

.env

+gemini_api_key="AIzaSyBrXveSu74dXPccUXcYGq0ZtmaiMCnoXyE"

agents.py

+import google.generativeai as genai
+import yfinance as yf
+from dotenv import load_dotenv
+
+
+load_dotenv()
+genai.configure(api_key="gemini_api_key")
+
+def analyze_stock(symbol):
+    stock = yf.Ticker(symbol)
+    hist = stock.history(period="1mo")
+
+    if hist.empty:
+        return f"No data found for symbol '{symbol}'. Please check the ticker."
+
+    avg_price = hist['Close'].mean()
+    last_price = hist['Close'].iloc[-1]
+    trend = "upward 📈" if last_price > avg_price else "downward 📉"
+
+    # LLM enhancement
+    prompt = f"""
+    Provide a short stock market analysis for {symbol} based on this info:
+    - 1 month average closing price: {avg_price:.2f}
+    - Last closing price: {last_price:.2f}
+    - Trend: {trend}
+    Keep it concise and professional.
+    """
+    model = genai.GenerativeModel("gemini-2.5-flash")
+    llm_response = model.generate_content(prompt)
+
+    return llm_response.text
+
+def risk_assessment(symbol):
+    stock = yf.Ticker(symbol)
+    hist = stock.history(period="6mo")['Close']
+    volatility = hist.pct_change().std() * (252 ** 0.5)
+    risk_level = "High ⚠️" if volatility > 0.3 else "Moderate 🟡" if volatility > 0.15 else "Low ✅"
+
+    # LLM enhancement
+    prompt = f"""
+    Explain the risk for {symbol} based on this data:
+    - 6-month volatility: {volatility:.2f}
+    - Risk level: {risk_level}
+    Provide practical guidance for an investor in plain English.
+    """
+    model = genai.GenerativeModel("gemini-2.5-flash")
+    llm_response = model.generate_content(prompt)
+
+    return llm_response.text
+def compare_stocks(symbol1, symbol2):
+    stock1 = yf.Ticker(symbol1)
+    stock2 = yf.Ticker(symbol2)
+
+    hist1 = stock1.history(period="1mo")
+    hist2 = stock2.history(period="1mo")
+
+    if hist1.empty or hist2.empty:
+        missing = []
+        if hist1.empty: missing.append(symbol1)
+        if hist2.empty: missing.append(symbol2)
+        return f"No data found for {', '.join(missing)}. Please check the ticker(s)."
+
+    avg1, last1 = hist1['Close'].mean(), hist1['Close'].iloc[-1]
+    avg2, last2 = hist2['Close'].mean(), hist2['Close'].iloc[-1]
+
+    perf1 = (last1 - hist1['Close'].iloc[0]) / hist1['Close'].iloc[0] * 100
+    perf2 = (last2 - hist2['Close'].iloc[0]) / hist2['Close'].iloc[0] * 100
+
+    better = symbol1 if perf1 > perf2 else symbol2
+
+    prompt = f"""
+    Compare two stocks: {symbol1} and {symbol2}.
+    - {symbol1}: Avg = {avg1:.2f}, Last = {last1:.2f}, 1-month change = {perf1:.2f}%
+    - {symbol2}: Avg = {avg2:.2f}, Last = {last2:.2f}, 1-month change = {perf2:.2f}%
+    Suggest which stock is performing better and why.
+    """
+    model = genai.GenerativeModel("gemini-2.5-flash")
+    llm_response = model.generate_content(prompt)
+
+    return llm_response.text 
+

app.py

+import streamlit as st
+import google.generativeai as genai
+from dotenv import load_dotenv
+import os
+from router import agent_router  
+
+# Load API key
+load_dotenv()
+genai.configure(api_key=os.getenv("gemini_api_key"))
+
+# Streamlit UI setup
+st.set_page_config(page_title="Stock Market Consultant 💹", layout="wide")
+st.title("💬 Stock Market Consultant")
+
+# Initialize chat history
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
+# --- Chat container: show messages ABOVE the input ---
+chat_container = st.container()
+
+# Display previous chat messages (top to bottom)
+with chat_container:
+    for msg in st.session_state.messages:
+        with st.chat_message(msg["role"]):
+            st.markdown(msg["content"])
+
+# --- Input area stays at the BOTTOM ---
+st.markdown("---")  # separator line for clarity
+
+with st.container():
+    with st.form(key="chat_form", clear_on_submit=True):
+        user_input = st.text_input(
+            "💬 Type your question about stocks or crypto:",
+            placeholder="e.g., search for analysing,risk factors and comparing of stocks",
+            label_visibility="collapsed"
+        )
+        submitted = st.form_submit_button("Send")
+
+    if submitted and user_input:
+        # Display user message
+        with st.chat_message("user"):
+            st.markdown(user_input)
+        st.session_state.messages.append({"role": "user", "content": user_input})
+
+        # Get AI/agent response
+        with st.chat_message("assistant"):
+            with st.spinner("Analyzing market data..."):
+                response_text = agent_router(user_input)
+                st.markdown(response_text)
+
+        # Save assistant message
+        st.session_state.messages.append({"role": "assistant", "content": response_text})

read.md

+# 💹 Agentic AI Stock Market Consultant
+
+An AI-powered system that performs stock market trend, risk, and comparison analysis using **Gemini API** and **Agentic AI** principles.
+
+## 🧠 Overview
+This project demonstrates an intelligent, multi-agent stock consultant system that autonomously analyzes and compares stocks using financial data from Yahoo Finance and reasoning from Google Gemini LLM.
+
+## ⚙️ Features
+- 💬 Conversational chat interface using Streamlit  
+- 📈 Trend analysis for stock prices  
+- ⚠️ Risk assessment based on volatility  
+- 🧮 Stock performance comparison  
+- 🧠 LLM-driven decision-making (Gemini API)
+
+## 🧩 System Architecture
+User Query → Router Agent  → Gemini Response → Specialized Stock Agents → Gemini Response → Streamlit Chat UI
+
+## 🧠 Agentic AI Design
+- Router Agent → decides which analysis tool to use  
+- Analysis Agent → analyzes price trends  
+- Risk Agent → calculates volatility  
+- Comparison Agent → compares two stocks  
+
+## 🧰 Tech Stack
+- Python 3.10+
+- Streamlit
+- yFinance
+- Google Gemini API
+- dotenv
+
+## 🚀 Setup
+1. Clone the repo  
+2. Create a `.env` file with your API key  
+3. Install dependencies:  
+   ```bash
+   pip install -r requirements.txt
+4.streamlit run app.py
+
+
+---
+
+### **📁 Folder Structure**
+
+Show your project organization.
+
+```markdown
+## 📁 Folder Structure
+project/
+├── app.py                 # Main Streamlit app
+├── router.py              # Routes queries to the appropriate agent
+├── agents.py              # Stock analysis, risk, and comparison functions
+├── resolve_ticker.py      # LLM-based ticker symbol extraction
+├── .env                   # Environment variables (API keys)
+├── requirements.txt       # Python dependencies
+└── README.md              # Project documentation
+
+
+## 🧮 Example Queries
+| User Query | Action Taken |
+|-------------|--------------|
+| "Analyze AAPL" | Trend analysis |
+| "Risk TSLA" | Risk agent |
+| "Compare NVDA and AMD" | Comparison agent |
+
+## 📈 Future Enhancements
+- Add sentiment analysis using financial news  
+- Implement portfolio optimization  
+- Add voice-based Gemini interaction
+
+## 🏆 Conclusion
+This project demonstrates an Agentic AI system that autonomously decides, reasons, and executes stock market analysis using Gemini API and custom agents.
+
+## 👨‍💻 Author 
+Name: Allen Bose
+Year: 2025
+
+
+
+
+

resolve_ticker.py

+import google.generativeai as genai
+
+def resolve_tickers(query):
+    """
+    Uses Gemini LLM to find stock ticker symbols mentioned in natural text.
+    Example: "compare Tesla and Apple" → ["TSLA", "AAPL"]
+    """
+    model = genai.GenerativeModel("gemini-2.5-flash")
+    prompt = f"""
+    Extract the official Yahoo Finance ticker symbols for the companies mentioned in this text:
+    "{query}"
+    Respond with only a comma-separated list of ticker symbols (like TSLA,AAPL,NVDA).
+    """
+    response = model.generate_content(prompt)
+    tickers = [t.strip().upper() for t in response.text.split(",") if t.strip()]
+    return tickers
+
+import google.generativeai as genai
+
+def detect_intent(query):
+    """
+    Uses Gemini LLM to understand user intent and return one of:
+    'analyze', 'risk', 'compare', or 'general'
+    """
+    model = genai.GenerativeModel("gemini-2.5-flash")
+    prompt = f"""
+    You are a routing assistant in a financial analysis system.
+    Analyze the following user query and classify it into one of four categories:
+    - 'analyze' → if user wants to check stock performance, trend, or price movement
+    - 'risk' → if user asks about volatility, safety, or risk level
+    - 'compare' → if user compares two or more companies
+    - 'general' → if it’s general finance or unrelated
+
+    Query: "{query}"
+
+    Respond with only one word: analyze, risk, compare, or general.
+    """
+    response = model.generate_content(prompt)
+    return response.text.strip().lower()

router.py

+from agents import analyze_stock, risk_assessment, compare_stocks
+from resolve_ticker import resolve_tickers, detect_intent
+import google.generativeai as genai
+
+def agent_router(query):
+    """
+    Routes the user query to the appropriate function or LLM.
+    Ensures responses are concise and focused.
+    """
+    intent = detect_intent(query)
+    tickers = resolve_tickers(query)
+
+    # Handle missing tickers
+    if not tickers:
+        return "⚠️ Couldn’t find any valid stock symbol. Please check your input."
+
+    # Route based on intent
+    if intent == "analyze":
+        raw_result = analyze_stock(tickers[0])
+        return raw_result
+    elif intent == "risk":
+        raw_result = risk_assessment(tickers[0])
+        return raw_result
+    elif intent == "compare" and len(tickers) >= 2:
+        raw_result = compare_stocks(tickers[0], tickers[1])
+        return raw_result
+    else:
+        prompt = f"""
+        Respond concisely to the user's query in 2-3 sentences, only providing the essential info. 
+        Query: {query}
+        """
+        model = genai.GenerativeModel("gemini-2.5-flash")
+        response = model.generate_content(prompt)
+        return response.text
+    
\ No newline at end of file