Upload folder using huggingface_hub

README.md

@@ -1,94 +1,28 @@
 ---
-title: Stock Analyst
-emoji: 📈
-colorFrom: blue
-colorTo: green
+title: Midas
+emoji: 🪙
+colorFrom: indigo
+colorTo: yellow
 sdk: docker
-sdk_version: "0.0.1"
 app_file: app.py
 pinned: false
 license: mit
-short_description: Multi-Agent Stock Analysis Team (uses AutoGen)
+short_description: Multi-agent investment team for stock analysis & decisions
 ---
 
-# Stock Analyst
+# Midas
 
-This is an experimental multi-agent system for comprehensive stock market analysis. It uses **Microsoft AutoGen** to orchestrate a team of specialized agents that collaborate to provide deep insights and investment decisions.
+Midas — the king with the golden touch — is a multi-agent investment analysis team. Four specialist agents evaluate a stock from every angle and converge on a final investment recommendation.
 
-## Features
-- **Team of Agents**: Collaborative analysis from Trends, News, Sentiment, and Decision agents.
-- **Round-Robin Orchestration**: Agents take turns sharing insights in a structured conversation.
-- **Real-time Data**: Fetches live stock history and financial data via `yfinance`.
-- **News Integration**: Searches DuckDuckGo for the latest market news.
-- **Streamlit UI**: Clean, interactive interface with agent avatars and real-time streaming.
+## What it does
 
-## Usage
-1. Enter a valid stock ticker (e.g., TSLA, NVDA, AAPL).
-2. Click **Analyze Stock**.
-3. Watch the agents collaborate in real-time to analyze trends, news, and sentiment alongside the "Decision Agent" making the final call.
+- **Trend Agent** — price history, moving averages, momentum signals
+- **News Agent** — latest headlines and sentiment for the ticker
+- **Sentiment Agent** — quantified market mood from news and social signals
+- **Decision Agent** — synthesises all inputs into a Buy / Hold / Sell verdict
 
-## Supported Tools
-- **Yahoo Finance**: Historical prices, analyst recommendations, market sentiment.
-- **DuckDuckGo**: Live news search.
-- **Web Scraping**: Fetching and summarizing full article content.
+## Stack
 
-## Project Folder Structure
+Streamlit · OpenAI Agents SDK · Yahoo Finance · Google Gemini / GPT-4o
 
-```
-stock-analyst/
-├── app.py                    # Main Streamlit UI
-├── aagents/
-│   ├── agents.py             # Agent definitions (Trends, News, Sentiment, Decision) and factories
-├── teams/
-│   ├── team.py               # RoundRobinGroupChat orchestration logic
-├── tools/
-│   ├── yf_tools.py           # Yahoo Finance tool wrappers
-│   ├── search_tools.py       # DuckDuckGo and web scraping tools
-├── Dockerfile                # Deployment configuration
-└── README.md                 # Project documentation
-```
-
-## Agents (`aagents/agents.py`)
-
-- **Stock Trends Agent** (📈):
-  - Fetches historical price data.
-  - Analyzes price movements and volume trends.
-  - Outputs structured `StockTrend` data for UI visualization.
-
-- **News Agent** (📰):
-  - Searches for top recent news stories.
-  - Fetches and reads full article content.
-  - Summarizes key events impacting the stock.
-
-- **Sentiment Agent** (💡):
-  - Check general market sentiment.
-  - Reviews analyst recommendations.
-  - Aggregates expert opinions.
-
-- **Decision Agent** (⚖️):
-  - Synthesizes all gathered information.
-  - Provides a final "Invest" or "Not Invest" decision.
-  - Summarizes the rationale.
-
-## Key Technologies
-
-| Component | Technology | Purpose |
-|-----------|-----------|---------|
-| Agent Framework | Microsoft AutoGen | Multi-agent orchestration |
-| LLM | GPT-4o / Gemini | Intelligence engine for agents |
-| UI Framework | Streamlit | User interface |
-| Data Source | yfinance | Stock market data |
-| Search | DuckDuckGo | Real-time news |
-
-## Running Locally
-
-```bash
-# Install dependencies
-uv sync
-
-# Set environment variables in .env or shell
-export GOOGLE_API_KEY="your-gemini-key" # or OPENAI_API_KEY if using OpenAI
-
-# Run the Streamlit app (from the root)
-streamlit run src/stock-analyst/app.py
-```
+> **Disclaimer:** For informational purposes only. Not financial advice.

pyproject.toml

@@ -1,5 +1,5 @@
 [project]
-name = "agents"
+name = "agenticai"
 version = "0.1.0"
 description = "Agentic AI project"
 readme = "README.md"
@@ -12,128 +12,156 @@ dependencies = [
     "openai>=2.8.1",
     "openai-agents>=0.5.1",
     "anthropic>=0.49.0",
-    "langchain-openai>=1.0.3",
+    "langchain-openai>=1.1.14",
     "langchain-anthropic>=1.1.0",
     "langchain_huggingface>=1.1.0",
     "langchain_ollama>=1.0.0",
     "langchain_google_genai>=3.0.3",
     "langchain_groq>=1.0.1",
-
-
     # =======================
     # LANGCHAIN / LANGGRAPH
     # =======================
     "langchain>=1.0.7",
     "langchain-community>=0.4.1",
-    "langgraph>=1.0.3",
+    "langgraph>=1.0.10rc1",
     "langgraph-checkpoint-sqlite>=3.0.0",
-    "langsmith>=0.4.43",
-    "langchain-text-splitters>=1.0.0",
+    "langsmith>=0.8.0",
+    "langchain-text-splitters>=1.1.2",
     "langchain-chroma>=1.0.0",
     "html2text>=2025.4.15",
     "traceloop-sdk>=0.33.0",
-
+    # ==================================
+    #  LlamIndex
+    # ==================================
+    "llama-index>=0.14.15",
+    "llama-index-llms-google-genai>=0.8.7",
+    "llama-index-embeddings-google-genai>=0.3.2",
+    "google-generativeai>=0.8.6",
+    # =======================
+    # MICROSOFT AGENT FRAMEWORK
+    # =======================
+    #"agent-framework==1.0.0b251204",
+    #"agent-framework-azure-ai==1.0.0b251204",
+    #"azure-ai-projects",
+    #"azure-ai-agents",
+    #"azure-ai-agents>=1.2.0b5",
+    #"agent-framework-azure-ai",
     # =======================
     # VECTOR DB / INDEXING
     # =======================
     "faiss-cpu>=1.13.0",
-    "chromadb==1.3.5",
+    "chromadb>=0.4.0",
     "sentence-transformers>=5.1.2",
     "pymupdf",
-    "pypdf>=6.3.0",
+    "pypdf>=6.10.2",
     "pypdf2>=3.0.1",
     "arxiv>=2.3.1",
     "wikipedia>=1.4.0",
-
     # =======================
     # AUTOGEN
     # =======================
-    "autogen-agentchat>=0.7.5",
-    "autogen-ext[grpc,mcp,ollama,openai]>=0.7.5",
+    "autogen-agentchat==0.7.5",
+    "autogen-ext[grpc,mcp,ollama,openai]==0.7.5",
     "asyncio",
-
+    "phidata>=2.0.0",
     # =======================
     # MCP
     # =======================
     "mcp-server-fetch>=2025.1.17",
     "mcp[cli]>=1.21.2",
-    
     # =======================
     # NETWORKING / UTILITIES
     # =======================
     "psutil>=7.0.0",
-    "python-dotenv>=1.0.1",
-    "requests>=2.32.3",
-    "aiohttp>=3.8.5",
+    "python-dotenv>=1.2.2",
+    "requests>=2.33.0",
+    "aiohttp>=3.14.0",
     "httpx>=0.28.1",
     "speedtest-cli>=2.1.3",
     "logfire",
     "google-search-results",
     "smithery>=0.4.4",
     "sendgrid",
-
     # =======================
     # WEB SCRAPING
     # =======================
     "playwright>=1.51.0",
     "beautifulsoup4>=4.12.3",
-    "lxml>=5.3.1",
-
+    "lxml>=6.1.0",
     # =======================
     # FINANCE / NLP
     # =======================
     "yfinance>=0.2.66",
     "textblob>=0.17.1",
     "polygon-api-client>=1.16.3",
-
     # =======================
     # VISUAL / UI / PDF
     # =======================
     "plotly>=6.5.0",
-    "streamlit>=1.51.0",
+    "streamlit>=1.54.0",
     "reportlab>=4.4.5",
     "fastapi",
-
+    "Pillow",
+    "python-docx",
+    "matplotlib",
+    "fpdf",
+    "extra-streamlit-components",
+    "nest_asyncio",
     # =======================
     # AUDIO / VIDEO
     # =======================
-    "yt_dlp>=2025.11.12",
-    "openai-whisper>=1.0.0",
-
+    "yt_dlp>=2026.2.21",
+    "openai-whisper==20250625",
+    "numba==0.63.1",
+    "llvmlite==0.46.0",
     # =======================
     # MACHINE LEARNING
     # =======================
     "scikit-learn>=1.7.2",
-    "huggingface_hub<=1.1.4",
+    "huggingface_hub>=0.23.2",
     "datasets>=4.4.1",
-
     # =======================
     # IPYNB SUPPORT
     # =======================
     "ipykernel>=7.1.0",
-
     # =======================
     # TOOLS
     # =======================
     "ddgs>=9.9.2",
     "duckduckgo_search",
     "azure-identity>=1.25.1",
-
+    "azure-mgmt-resource>=23.0.1",
+    "azure-mgmt-compute>=30.3.0",
+    "azure-mgmt-monitor>=6.0.2",
+    "azure-monitor-query>=1.2.0",
+    "PyGithub>=2.1.1",
     # =======================
     # OBSERVABILITY
     # =======================
+    "langfuse>=3.0.0",
     "openinference-instrumentation-autogen>=0.1.0",
     "openinference-instrumentation-openai>=0.1.15",
     "opentelemetry-sdk>=1.20.0",
     "opentelemetry-exporter-otlp>=1.20.0",
     "opentelemetry-api>=1.20.0",
+    # =======================
+    # Google Authentication
+    # =======================
+    "google-auth>=2.22.0",
+    "google-auth-oauthlib>=0.4.6",
+    "google-auth-httplib2>=0.1.0",
+    "autoflake>=1.5.0",
+    "psycopg2-binary>=2.9.9",
+    "sqlalchemy>=2.0.46",
+    "llama-index-readers-database>=0.5.1",
 ]
 
 [dependency-groups]
 dev = [
-    "pytest>=8.3.3",
+    "pytest>=9.0.3",
     "ipykernel>=7.1.0",
     "pytest-asyncio",
+    "pytest-cov>=7.1.0",
 ]
 
 # ============================================================
@@ -149,20 +177,16 @@ build-backend = "setuptools.build_meta"
 # PACKAGING & DISCOVERY
 # ============================================================
 # Tells setuptools where to find the source code.
-# This makes 'common' and 'src' importable when installed (pip install -e .).
 [tool.setuptools.packages.find]
-where = ["."]  # Look in the project root
-include = ["common*", "src*"]  # Treat 'common' and 'src' folders as packages
+where = ["."]
+include = ["src*"]
 
 
 # ============================================================
 # PYTEST SETTINGS
 # ============================================================
-# Configures the test runner to automatically find code.
 [tool.pytest.ini_options]
-# Adds 'src' and 'common' to the python path during tests.
-# This allows tests to import modules (e.g., 'import travel_agent')
-# just like the apps do locally, preventing ModuleNotFoundError.
-pythonpath = ["src", "common"]
+pythonpath = ["src"]
 testpaths = ["tests"]  # Only look for tests in the 'tests' directory
 addopts = "-q"  # Run in quiet mode (less verbose output)
+asyncio_mode = "auto"  # Auto-detect async tests (requires pytest-asyncio)

run.py

@@ -16,9 +16,10 @@ import sys
 import os
 import subprocess
 import argparse
+import time
 from pathlib import Path
 from typing import Dict, Optional
-from agents import Runner, SQLiteSession
+# from agents import Runner, SQLiteSession
 # from agents import set_trace_processors
 # from langsmith.wrappers import OpenAIAgentsTracingProcessor
 
@@ -28,46 +29,49 @@ load_dotenv(override=True)
 
 # App registry - maps app names to their paths and entry points
 APP_REGISTRY: Dict[str, Dict[str, str]] = {
-    "healthcare": {
-        "path": "src/healthcare-assistant",
+    "research-assistant": {
+        "path": "src/ai-research-assistant",
         "entry": "app.py",
-        "description": "Healthcare Assistant - Medical information with RAG and web search"
+        "description": "AI Research Assistant - Multi-specialist orchestrator for finance, news, and web research"
     },
     "deep-research": {
-        "path": "src/deep-research",
+        "path": "src/deep-research-reporter",
+        "entry": "app.py",
+        "description": "Deep Research Reporter - Plans, searches, and synthesises comprehensive research reports"
+    },
+    "healthcare": {
+        "path": "src/healthcare-rag-advisor",
         "entry": "app.py",
-        "description": "Deep Research AI - Comprehensive research assistant"
+        "description": "Healthcare RAG Advisor - Medical information retrieval using RAG and web search"
     },
     "stock-analyst": {
-        "path": "src/stock-analyst",
+        "path": "src/stock-investment-analyst",
         "entry": "app.py",
-        "description": "Stock Analyst - Financial analysis and stock recommendations"
+        "description": "Stock Investment Analyst - Multi-agent investment team for technical and sentiment analysis"
     },
-    "travel-agent": {
-        "path": "src/travel-agent",
+    "travel-planner": {
+        "path": "src/travel-planner",
         "entry": "app.py",
-        "description": "Travel Agent - Trip planning and travel recommendations"
+        "description": "Travel Planner - AI-powered trip planning with flight, hotel, and itinerary recommendations"
     },
-    "trip-planner": {
-        "path": "src/trip-planner",
+    "trip-planner-api": {
+        "path": "src/trip-planner-api",
         "entry": "main.py",
-        "description": "Trip Planner - Detailed trip itinerary planning"
+        "type": "fastapi",
+        "description": "Trip Planner API - Phidata-powered trip itinerary planning REST API"
     },
-    "chatbot": {
-        "path": "src/chatbot",
-        "entry": "app.py",
-        "description": "General Chatbot - Multi-purpose conversational AI"
+    "market-analyst": {
+        "path": "src/market-analyst",
+        "entry": "backend/main.py",
+        "type": "fastapi",
+        "description": "AI Market Analyst - Real-time multi-agent market analysis with streaming (Vue.js + FastAPI)"
     },
-    "accessibility": {
-        "path": "src/accessibility",
-        "entry": "app.py",
-        "description": "Accessibility Tools - Assistive technology applications"
+    "test": {
+        "path": ".",
+        "entry": "tests",
+        "type": "test",
+        "description": "Run Project Tests - Executes pytest suite"
     },
-    "literature-review": {
-        "path": "src/literature-review",
-        "entry": "app.py",
-        "description": "Literature Review Assistant - Multi-agent literature review tool"
-    }
 }
 
 
@@ -142,33 +146,91 @@ def launch_app(app_name: str, port: Optional[int] = None):
     print(f"📂 Location: {config['path']}")
     print(f"🌐 Entry Point: {app_file}")
     
-    # Build streamlit command
-    cmd = ["streamlit", "run", app_file]
+    app_type = config.get("type", "streamlit")
     
-    # Add port if specified
-    if port:
-        cmd.extend(["--server.port", str(port)])
-        print(f"🔌 Port: {port}")
+    python_exe = sys.executable
+    is_windows = sys.platform == "win32"
+    
+    # Prepare environment with project root in PYTHONPATH to fix imports
+    env = os.environ.copy()
+    env["PYTHONPATH"] = str(project_root) + os.pathsep + env.get("PYTHONPATH", "")
+    
+    # Decoupled App Logic: Build frontend if needed
+    if app_name == "market-analyst":
+        frontend_dir = project_root / "src/market-analyst/frontend"
+        dist_dir = frontend_dir / "dist"
+        if not dist_dir.exists():
+            print("\n🛠️ Frontend build missing. Building now...")
+            subprocess.run(["npm", "run", "build"], cwd=frontend_dir, shell=is_windows)
+            print("✅ Frontend built.\n")
+    
+    # App Type specific logic
+    if app_type == "fastapi":
+        # Extract module name from entry point (e.g. backend/main.py -> backend.main)
+        module_path = app_file.replace(".py", "").replace("/", ".").replace("\\", ".")
+        cmd = [python_exe, "-m", "uvicorn", f"{module_path}:app", "--host", "0.0.0.0"]
+        default_port = 8000
+    elif app_type == "script":
+        cmd = [python_exe, app_file]
+        default_port = None
+    elif app_type == "test":
+        cmd = [python_exe, "-m", "pytest", app_file, "-v"]
+        default_port = None
+    elif app_type == "npm":
+        cmd = ["npm", "run", "dev"]
+        default_port = 5173
     else:
-        print(f"🔌 Port: 8501 (default)")
+        cmd = [python_exe, "-m", "streamlit", "run", app_file]
+        default_port = 8501
+    
+    # Add port if specified
+    actual_port = port if port else default_port
+    
+    if app_type in ["streamlit", "fastapi", "npm"]:
+        if port:
+            if app_type == "fastapi":
+                cmd.extend(["--port", str(port)])
+            elif app_type == "npm":
+                cmd.extend(["--", "--port", str(port)])
+            else:
+                cmd.extend(["--server.port", str(port)])
+            print(f"🔌 Port: {port}")
+        else:
+            print(f"🔌 Port: {default_port} (default)")
+            
+        # Kill any process using the target port (Port is only relevant for web apps)
+        try:
+            import platform
+            if platform.system() != "Windows":
+                # Use fuser on Linux/Mac to kill processes on the port
+                kill_cmd = ["fuser", "-k", f"{actual_port}/tcp"]
+                subprocess.run(kill_cmd, stderr=subprocess.DEVNULL, stdout=subprocess.DEVNULL)
+                print(f"🧹 Cleaned up port {actual_port}")
+        except Exception:
+            pass  # Silently continue if cleanup fails
     
     print("\n" + "=" * 70)
     print("\n🎯 Starting application...\n")
     
-    # Prepare environment with project root in PYTHONPATH to fix imports
-    env = os.environ.copy()
-    env["PYTHONPATH"] = str(project_root) + os.pathsep + env.get("PYTHONPATH", "")
     print(f"\n\nPYTHONPATH: {env['PYTHONPATH']}")
 
     try:
         # Change to app directory and run
         os.chdir(app_dir)
-        subprocess.run(cmd, env=env)
+        
+
+
+        subprocess.run(cmd, env=env, shell=is_windows)
     except KeyboardInterrupt:
         print("\n\n👋 Application stopped by user")
     except FileNotFoundError:
-        print("\n❌ Error: Streamlit not found. Please install it:")
-        print("   pip install streamlit")
+        binary = "command"
+        if app_type == "fastapi": binary = "uvicorn"
+        elif app_type == "streamlit": binary = "streamlit"
+        elif app_type == "test": binary = "pytest"
+        
+        print(f"\n❌ Error: {binary} not found in the current environment.")
+        print(f"   Please install it: pip install {binary}")
         sys.exit(1)
     except Exception as e:
         print(f"\n❌ Error launching app: {e}")

src/stock-investment-analyst/Dockerfile

@@ -0,0 +1,36 @@
+FROM python:3.12-slim
+
+ENV PYTHONUNBUFFERED=1 \
+    DEBIAN_FRONTEND=noninteractive \
+    PYTHONPATH=/app/src/stock-analyst:$PYTHONPATH
+
+WORKDIR /app
+
+# System deps
+RUN apt-get update && apt-get install -y \
+    git build-essential curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install uv
+RUN curl -LsSf https://astral.sh/uv/install.sh | sh
+ENV PATH="/root/.local/bin:$PATH"
+
+# Copy project metadata
+COPY pyproject.toml .
+COPY uv.lock .
+
+# Copy application code
+COPY common/ ./common/
+COPY src/stock-analyst/ ./src/stock-analyst/
+
+# Install dependencies using uv, then export and install with pip to system
+# We use --no-dev to exclude dev dependencies if any
+RUN uv sync --frozen --no-dev && \
+    uv pip install -e . --system
+
+# Copy entry point
+COPY run.py .
+
+EXPOSE 7860
+
+CMD ["python", "run.py", "stock-analyst", "--port", "7860"]

src/stock-investment-analyst/README.md

@@ -0,0 +1,28 @@
+---
+title: Midas
+emoji: 🪙
+colorFrom: indigo
+colorTo: yellow
+sdk: docker
+app_file: app.py
+pinned: false
+license: mit
+short_description: Multi-agent investment team for stock analysis & decisions
+---
+
+# Midas
+
+Midas — the king with the golden touch — is a multi-agent investment analysis team. Four specialist agents evaluate a stock from every angle and converge on a final investment recommendation.
+
+## What it does
+
+- **Trend Agent** — price history, moving averages, momentum signals
+- **News Agent** — latest headlines and sentiment for the ticker
+- **Sentiment Agent** — quantified market mood from news and social signals
+- **Decision Agent** — synthesises all inputs into a Buy / Hold / Sell verdict
+
+## Stack
+
+Streamlit · OpenAI Agents SDK · Yahoo Finance · Google Gemini / GPT-4o
+
+> **Disclaimer:** For informational purposes only. Not financial advice.

src/stock-investment-analyst/aagents/__init__.py

@@ -0,0 +1,16 @@
+from .stock_trends_agent import get_stock_trends_agent
+from .news_agent import get_news_agent
+from .sentiment_agent import get_sentiment_agent
+from .decision_agent import get_decision_agent
+from pydantic import BaseModel
+
+class StockTrend(BaseModel):
+    stock_name: str
+    trade_date: str
+    open_price: float
+    close_price: float
+    high_price: float
+    low_price: float
+    volume: int
+
+__all__ = ["get_stock_trends_agent", "get_news_agent", "get_sentiment_agent", "get_decision_agent", "StockTrend"]

src/stock-investment-analyst/aagents/decision_agent.py

@@ -0,0 +1,50 @@
+from autogen_agentchat.agents import AssistantAgent
+from core.model import get_model_client
+
+def get_decision_agent():
+    # Upgrade to Pro model for better reasoning
+    # Using 1.5-pro as requested for better decision making
+    model_client = get_model_client()
+
+    decision_agent = AssistantAgent(
+        name="decision_agent",
+        model_client=model_client,
+        tools=[], # Decision agent usually synthesizes info, might not need tools if it consumes chat history
+        system_message=(
+            "You are the Decision Agent. Your role is to synthesize data from the Stock Trends, News, and Sentiment agents "
+            "to provide a final, well-reasoned investment recommendation.\n\n"
+            
+            "**STEP 1: CALCULATE WEIGHTED SCORE**\n"
+            "You MUST score the stock on the following criteria (0-10 scale) and calculate the weighted total:\n"
+            "1. **Technical Indicators (Weight: 40%)**: Score 0-10 based on trend direction, moving averages, and volume.\n"
+            "2. **News Sentiment (Weight: 30%)**: Score 0-10 based on recent headlines and PR tone.\n"
+            "3. **Analyst Ratings (Weight: 30%)**: Score 0-10 based on analyst consensus and price targets.\n\n"
+            
+            "**Formula**: `(Technical * 0.4) + (News * 0.3) + (Analyst * 0.3) = Total Score`\n\n"
+            
+            "**STEP 2: DETERMINE DECISION**\n"
+            "- If **Total Score > 7.5** -> Decision: **INVEST**\n"
+            "- If **Total Score < 5.0** -> Decision: **AVOID**\n"
+            "- Else -> Decision: **WAIT**\n\n"
+
+            "**Output Requirement:**\n"
+            "You MUST provide your response in the following structured format:\n"
+            "1. **Decision**: [Invest / Wait / Avoid] (Based strictly on the rule above)\n"
+            "2. **Scoring Table**:\n"
+            "   | Category | Score (0-10) | Weighted Score |\n"
+            "   | :--- | :--- | :--- |\n"
+            "   | Technicals (40%) | [Score] | [Val] |\n"
+            "   | News (30%) | [Score] | [Val] |\n"
+            "   | Analysts (30%) | [Score] | [Val] |\n"
+            "   | **TOTAL** | | **[Total Score]** |\n"
+            "3. **Risk Level**: [Low / Medium / High]\n"
+            "4. **Reasoning**:\n"
+            "   - **Pros**: [List top 3 positive factors]\n"
+            "   - **Cons**: [List top 3 negative factors]\n"
+            "5. **Validation**: Briefly explain why the confidence score was chosen based on the consistency of the data.\n\n"
+            
+            "Also provide the current stock price if available in the context.\n"
+            "End your response with 'Decision Made' once you finalize the decision."
+        )
+    )
+    return decision_agent

src/stock-investment-analyst/aagents/news_agent.py

@@ -0,0 +1,38 @@
+from autogen_agentchat.agents import AssistantAgent
+from autogen_core.tools import FunctionTool 
+from tools.search_tools import _duckduckgo_search, searchQuery
+from core.model import get_model_client
+
+def get_news_agent():
+    model_client = get_model_client()
+
+    async def news_search(query: str) -> str:
+        """
+        Search for latest news regarding a topic or stock.
+        """
+        # Anchor search to reputable sources as per Suggestion 3
+        # reputable_sources = " (site:bloomberg.com OR site:reuters.com OR site:cnbc.com OR site:wsj.com OR site:finance.yahoo.com)"
+        # if "site:" not in query:
+        #      query += reputable_sources
+
+        # Use underlying function with proper params
+        params = searchQuery(query=query, search_type="news", timelimit="w", max_results=5)
+        # _duckduckgo_search returns list[dict], convert to str
+        results = _duckduckgo_search(params)
+        return str(results)
+
+    news_tool = FunctionTool(news_search, description="Search for latest top 5 news for a given stock or topic. Returns headlines and snippets only.")
+
+    news_agent = AssistantAgent(
+        name="news_agent",
+        model_client=model_client,
+        tools=[news_tool],
+        system_message=(
+            "You are the News Agent. "
+            "1. Search for the latest top 5 news stories related to the given stock using `news_tool`. "
+            "2. Prioritize reputable result sources like Bloomberg, Reuters, CNBC, WSJ, and Yahoo Finance if possible. "
+            "3. Summarize the key insights from the news stories. "
+            "Do NOT provide any final investment decision."
+        )
+    )
+    return news_agent

src/stock-investment-analyst/aagents/sentiment_agent.py

@@ -0,0 +1,32 @@
+from autogen_agentchat.agents import AssistantAgent
+from autogen_core.tools import FunctionTool 
+from tools.yf_tools import _get_analyst_recommendations
+from core.model import get_model_client
+
+def get_sentiment_agent():
+    model_client = get_model_client()
+
+    async def get_market_sentiment(symbol: str, period: str) -> str:
+        """Get market sentiment for a stock."""
+        from tools.yf_tools import _get_market_sentiment
+        return _get_market_sentiment(symbol, period)
+
+    async def get_analyst_recs(symbol: str) -> str:
+        """Get analyst recommendations for a stock."""
+        # _get_analyst_recommendations is already imported
+        return _get_analyst_recommendations(symbol)
+
+    sentiment_tool = FunctionTool(get_market_sentiment, description="Get market sentiment")
+    analyst_tool = FunctionTool(get_analyst_recs, description="Get analyst recommendations")
+
+    sentiment_agent = AssistantAgent(
+        name="sentiment_agent",
+        model_client=model_client,
+        tools=[sentiment_tool, analyst_tool],
+        system_message=(
+            "You are the Market Sentiment Agent. "
+            "You gather overall market sentiment, relevant analyst reports, and expert opinions. "
+            "Do NOT provide any final investment decision."
+        )
+    )
+    return sentiment_agent

src/stock-investment-analyst/aagents/stock_trends_agent.py

@@ -0,0 +1,34 @@
+from autogen_agentchat.agents import AssistantAgent
+from autogen_core.tools import FunctionTool 
+from tools.yf_tools import _get_history
+from core.model import get_model_client
+    
+def get_stock_trends_agent():
+    model_client = get_model_client()
+    
+    async def fetch_stock_history(symbol: str, period: str) -> str:
+        """
+        Gets real-time stock prices and changes over the last few months for the given stock name.
+        
+        Args:
+            symbol: The stock ticker symbol (e.g., 'TSLA', 'AAPL').
+            period: The period to fetch data for (e.g., '1mo', '3mo').
+        
+        Returns:
+            str: A formatted string showing the historical prices.
+        """
+        return _get_history(symbol, period)
+    
+    get_history_tool = FunctionTool(fetch_stock_history, description="Gets real-time stock prices, changes over the last few months for 'stock_name'", strict=True)
+    
+    stock_trends_agent_assistant = AssistantAgent(
+        name="stock_trends_agent",
+        model_client=model_client,
+        tools=[get_history_tool],
+        system_message=(
+            "You are the Stock Price Trends Agent practicing in India and USA stock markets. "
+            "You fetch and summarize stock prices, changes over the last 3 months, and general market trends. "
+            "Do NOT provide any final investment decision."
+        ),
+    )
+    return stock_trends_agent_assistant

src/stock-investment-analyst/app.py

@@ -0,0 +1,120 @@
+import streamlit as st
+import asyncio
+import os
+
+
+
+
+from teams.investment_team import get_investment_team
+
+st.set_page_config(page_title="Stock Investment Analyst", layout="wide", page_icon="📈")
+
+# ------------------------------------------------------------------------------
+# Custom CSS for layout improvements
+# ------------------------------------------------------------------------------
+st.markdown("""
+    <style>
+        .block-container {
+            padding-top: 1.5rem;
+            padding-bottom: 3rem;
+        }
+    </style>
+""", unsafe_allow_html=True)
+
+# ------------------------------------------------------------------------------
+# Main Content
+# ------------------------------------------------------------------------------
+st.title("📈 Stock Investment Analyst")
+
+# IMPORTANT DISCLAIMER
+st.warning(
+    "**⚠️ DISCLAIMER: EDUCATIONAL PURPOSE ONLY**\n\n"
+    "This tool is an AI-powered experiment designed for educational and demonstration purposes. "
+    "The analysis provided is generated by artificial intelligence and **DOES NOT** constitute financial, investment, or legal advice. "
+    "Do not use this tool to make real-world financial decisions. Always consult with a qualified financial advisor."
+)
+
+st.markdown("#### Get a comprehensive analysis from our AI team of agents.")
+
+# Input section with better spacing
+st.divider()
+with st.container():
+    col1, col2 = st.columns([3, 1], gap="medium")
+    with col1:
+        stock_name = st.text_input(
+            "Enter Stock Ticker", 
+            value="Tesla", 
+            placeholder="e.g. NVDA, TSLA, AAPL",
+            help="Enter the ticker symbol of the company you want to analyze."
+        )
+    with col2:
+        # Align button with input box
+        st.write("") # Spacer
+        st.write("") # Spacer
+        analyze_btn = st.button("🔍 Analyze Stock", type="primary", use_container_width=True)
+
+
+async def run_analysis(ticker):
+    # Start a span for the analysis task. 
+    # This becomes the parent for all subsequent spans (like OpenAI calls).
+    # with tracer.start_as_current_span("run_analysis") as span:
+    # span.set_attribute("stock.ticker", ticker) # Add useful metadata
+        
+    task = f"Analyze stock trends, news, and sentiment for {ticker}, plus analyst reports and expert opinions, and then decide whether to invest."
+    
+    st.markdown(f"### Analysis for **{ticker}**")
+    
+    # Container for live updates
+    chat_container = st.container()
+    
+    try:
+        # Run the team stream
+        investment_team = get_investment_team()
+        stream = investment_team.run_stream(task=task)
+        
+        # Define icons for each agent
+        AGENT_ICONS = {
+            "stock_trends_agent": "📈",
+            "news_agent": "📰",
+            "sentiment_agent": "💡",
+            "decision_agent": "⚖️",
+            "user": "👤",
+            "System": "⚙️"
+        }
+        
+        async for message in stream:
+            # Check if message has source and content attributes typical of agent messages
+            source = getattr(message, 'source', 'System')
+            
+            with chat_container:
+                if 'TaskResult' in message.__class__.__name__:
+                    if hasattr(message, 'stop_reason') and message.stop_reason:
+                            st.info(f"Analysis Completed: {message.stop_reason}")
+                    continue
+
+                # Use the icon mapping, default to None (Streamlit default) if not found
+                avatar = AGENT_ICONS.get(source, None)
+                
+                with st.chat_message(source, avatar=avatar):
+                    # Handle Tool Call events specifically to make them look like system logs
+                    if 'ToolCall' in message.__class__.__name__:
+                        with st.expander(f"⚙️ Tool Usage: {source}", expanded=False):
+                            st.write(message)
+                        continue
+
+                    content = getattr(message, 'content', "")
+                    st.write(content)
+                    
+    except Exception as e:
+        # Record the exception in the span if something crashes
+        # span.record_exception(e)
+        # span.set_status(trace.Status(trace.StatusCode.ERROR))
+        st.error(f"An error occurred during analysis: {e}")
+
+if analyze_btn:
+    if stock_name:
+        with st.spinner(f"Gathering data and analyzing {stock_name}..."):
+            # Create a new event loop for this run if needed, or simply run
+            asyncio.run(run_analysis(stock_name))
+    else:
+        st.warning("Please enter a valid stock ticker.")

src/stock-investment-analyst/core/__init__.py

@@ -0,0 +1,4 @@
+
+from .model import get_model_client
+
+__all__ = ["get_model_client"]

src/stock-investment-analyst/core/model.py

@@ -0,0 +1,32 @@
+from utility.autogen_model_factory import AutoGenModelFactory
+
+def get_model_client(provider:str = "google"):
+    if provider.lower() == "google":
+        return AutoGenModelFactory.get_model(
+            provider="google",
+            model_name="gemini-3-pro-preview",
+            temperature=0,
+        model_info={
+            "family": "gemini",
+            "vision": True,
+            "function_calling": True,
+            "json_output": True,
+            "structured_output": True,
+        },
+    )
+    elif provider.lower() == "openai":
+        return AutoGenModelFactory.get_model(
+            provider="openai",
+            model_name="gpt-4o-mini",
+            temperature=0,
+        model_info={
+            "family": "gpt",
+            "vision": True,
+            "function_calling": True,
+            "json_output": True,
+            "structured_output": True,
+        },
+    )
+    else: 
+        raise  ValueError(f"Unsupported provider: {provider}")
+

src/stock-investment-analyst/teams/__init__.py

@@ -0,0 +1,3 @@
+from .investment_team import get_investment_team
+
+__all__ = ["get_investment_team"]

src/stock-investment-analyst/teams/investment_team.py

@@ -0,0 +1,49 @@
+from autogen_agentchat.conditions import MaxMessageTermination, TextMentionTermination
+from autogen_agentchat.teams import SelectorGroupChat
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from autogen_agentchat.ui import Console
+import os
+
+from aagents import get_stock_trends_agent, get_news_agent, get_sentiment_agent, get_decision_agent, StockTrend
+from autogen_agentchat.messages import StructuredMessage
+from core.model import get_model_client
+
+def get_investment_team():
+    # Try to register the message type to avoid "not registered" errors in GroupChat
+    try:
+        from autogen_core import TypeSubscription
+        pass
+    except ImportError:
+        pass
+        
+    text_termination = TextMentionTermination("Decision Made")
+    max_message_termination = MaxMessageTermination(20)
+    termination = text_termination | max_message_termination
+    
+    # # Model for the selector/moderator
+    # selector_model = OpenAIChatCompletionClient(
+    #     model="gemini-2.5-flash",
+    #     api_key=os.getenv("GOOGLE_API_KEY"),
+    #     model_info={
+    #                     "family": "gemini",
+    #                     "vision": True,
+    #                     "function_calling": True,
+    #                     "json_output": True,
+    #                     "structured_output": True,
+    #                 },
+    #     temperature=0
+    # )
+
+    # Selector Group Chat which allows dynamic speaker selection
+    investment_team = SelectorGroupChat(
+        [
+            get_stock_trends_agent(),
+            get_news_agent(),
+            get_sentiment_agent(),
+            get_decision_agent(),
+        ],
+        model_client=get_model_client(),
+        termination_condition=termination
+    )
+    return investment_team
+

src/stock-investment-analyst/tools/__init__.py

@@ -0,0 +1,4 @@
+from .search_tools import _duckduckgo_search, searchQuery
+from .yf_tools import _get_history, _get_analyst_recommendations
+
+__all__ = ["_duckduckgo_search", "searchQuery", "_get_history", "_get_analyst_recommendations"]

src/stock-investment-analyst/tools/search_tools.py

@@ -0,0 +1,127 @@
+import requests
+from ddgs import DDGS
+from agents import function_tool
+
+from pydantic import BaseModel, Field
+from bs4 import BeautifulSoup
+from typing import Optional
+
+
+
+# ---------------------------------------------------------
+# Load environment variables
+# ---------------------------------------------------------
+
+
+# ---------------------- MODELS ---------------------------
+class searchQuery(BaseModel):
+    query: str = Field(..., description="The search query string.")
+    max_results: int = Field(5, description="The maximum number of search results to return.")
+    search_type: str = Field(
+        "text",
+        description="Search type: 'text' (default) or 'news'. Use 'news' to get publication dates."
+    )
+    timelimit: str = Field(
+        'd',
+        description="Time limit for search results: 'd' (day), 'w' (week), 'm' (month), 'y' (year)."
+    )
+    region: str = Field("us-en", description="Region for search results (e.g., 'us-en').")
+
+
+class searchResult(BaseModel):
+    title: str
+    link: str
+    snippet: str
+    datetime: Optional[str] = None
+
+
+# ---------------------- PAGE FETCH TOOL ---------------------------
+def _fetch_page_content(url: str, timeout: int = 3) -> Optional[str]:
+    """Fetch and extract text content from a web page."""
+    print(f"[DEBUG] fetch_page_content called with: {url} - timeout: {timeout}")
+    try:
+        headers = {
+            'User-Agent': (
+                'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
+                'AppleWebKit/537.36 (KHTML, like Gecko) '
+                'Chrome/91.0.4472.124 Safari/537.36'
+            )
+        }
+        response = requests.get(url, headers=headers, timeout=timeout)
+        response.raise_for_status()
+
+        soup = BeautifulSoup(response.content, 'html.parser')
+
+        # Remove irrelevant elements
+        for tag in soup(["script", "style", "nav", "footer", "header"]):
+            tag.decompose()
+
+        # Extract text
+        text = soup.get_text(separator='\n', strip=True)
+
+        # Clean whitespace
+        lines = (line.strip() for line in text.splitlines())
+        chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
+        text = '\n'.join(chunk for chunk in chunks if chunk)
+
+        return text
+    except Exception as e:
+        print(f"[WARNING] Failed to fetch content from {url}: {str(e)}")
+        return None
+
+
+@function_tool
+def fetch_page_content(url: str, timeout: int = 3) -> Optional[str]:
+    """Fetch and extract text content from a web page."""
+    return _fetch_page_content(url, timeout)
+
+
+# ---------------------- SEARCH TOOL ---------------------------
+def _duckduckgo_search(params: searchQuery) -> list[dict]:
+    """Perform a DuckDuckGo search and return only snippets.  
+    No page content fetched here."""
+    print(f"[DEBUG] duckduckgo_search called with: {params}")
+
+    results = []
+    with DDGS() as ddgs:
+        if params.search_type == "news":
+            search_results = ddgs.news(
+                params.query,
+                max_results=params.max_results,
+                timelimit=params.timelimit,
+                region=params.region
+            )
+            for result in search_results:
+                results.append(
+                    searchResult(
+                        title=result.get("title", ""),
+                        link=result.get("url", ""),
+                        snippet=result.get("body", ""),
+                        datetime=result.get("date", "")
+                    ).model_dump()
+                )
+        else:
+            search_results = ddgs.text(
+                params.query,
+                max_results=params.max_results,
+                timelimit=params.timelimit,
+                region=params.region
+            )
+            for result in search_results:
+                results.append(
+                    searchResult(
+                        title=result.get("title", ""),
+                        link=result.get("href", ""),
+                        snippet=result.get("body", "")
+                    ).model_dump()
+                )
+
+    print(f"[DEBUG] duckduckgo_search returning {len(results)} results")
+    return results
+
+@function_tool
+def duckduckgo_search(params: searchQuery) -> list[dict]:
+    """Perform a DuckDuckGo search and return only snippets.  
+    No page content fetched here."""
+    return _duckduckgo_search(params)
+

src/stock-investment-analyst/tools/yf_tools.py

@@ -0,0 +1,277 @@
+import os
+import requests
+import yfinance as yf
+
+from agents import function_tool
+from datetime import datetime, timedelta
+
+
+
+# Load environment variables
+
+
+
+# ============================================================
+# 🔹 YAHOO FINANCE TOOLSET
+# ============================================================
+def _get_summary(symbol: str, period: str = "1d", interval: str = "1h") -> str:
+    print(f"[DEBUG] get_summary called for symbol='{symbol}', period='{period}', interval='{interval}'")
+    try:
+        ticker = yf.Ticker(symbol)
+
+        # Calculate start and end dates based on period
+        end_date = datetime.today()
+        if period.endswith("d"):
+            days = int(period[:-1])
+        elif period.endswith("mo"):
+            days = int(period[:-2]) * 30
+        elif period.endswith("y"):
+            days = int(period[:-1]) * 365
+        else:
+            days = 30  # default 1 month
+        start_date = end_date - timedelta(days=days)
+
+        # Fetch recent data explicitly
+        data = ticker.history(
+            start=start_date.strftime("%Y-%m-%d"),
+            end=end_date.strftime("%Y-%m-%d"),
+            interval=interval
+        )
+
+        if data.empty:
+            return f"No data found for symbol '{symbol}'."
+
+        latest = data.iloc[-1]
+        current_price = round(latest["Close"], 2)
+        open_price = round(latest["Open"], 2)
+        change = round(current_price - open_price, 2)
+        pct_change = round((change / open_price) * 100, 2)
+
+        info = ticker.info
+        long_name = info.get("longName", symbol)
+        currency = info.get("currency", "USD")
+
+        formatted = [
+            f"📈 {long_name} ({symbol})",
+            f"Current Price: {current_price} {currency}",
+            f"Change: {change} ({pct_change}%)",
+            f"Open: {open_price} | High: {round(latest['High'], 2)} | Low: {round(latest['Low'], 2)}",
+            f"Volume: {int(latest['Volume'])}",
+            f"Period: {period} | Interval: {interval}",
+        ]
+        return "\n".join(formatted)
+
+    except Exception as e:
+        return f"Error fetching data for '{symbol}': {e}"
+
+def _get_market_sentiment(symbol: str, period: str = "1mo") -> str:
+    print(f"[DEBUG] get_market_sentiment called for symbol='{symbol}', period='{period}'")
+    try:
+        ticker = yf.Ticker(symbol)
+
+        # Calculate start/end dynamically
+        end_date = datetime.today()
+        if period.endswith("d"):
+            days = int(period[:-1])
+        elif period.endswith("mo"):
+            days = int(period[:-2]) * 30
+        elif period.endswith("y"):
+            days = int(period[:-1]) * 365
+        else:
+            days = 30
+        start_date = end_date - timedelta(days=days)
+
+        data = ticker.history(
+            start=start_date.strftime("%Y-%m-%d"),
+            end=end_date.strftime("%Y-%m-%d")
+        )
+
+        if data.empty:
+            return f"No data for {symbol}."
+
+        recent_change = data["Close"].iloc[-1] - data["Close"].iloc[0]
+        pct_change = (recent_change / data["Close"].iloc[0]) * 100
+
+        sentiment = "Neutral"
+        if pct_change > 2:
+            sentiment = "Bullish"
+        elif pct_change < -2:
+            sentiment = "Bearish"
+
+        return f"{symbol} market sentiment ({period}): {sentiment} ({pct_change:.2f}% change)"
+
+    except Exception as e:
+        return f"Error fetching market sentiment for '{symbol}': {e}"
+
+def _get_history(symbol: str, period: str = "1mo") -> str:
+    print(f"[DEBUG] get_history called for symbol='{symbol}', period='{period}'")
+    try:
+        ticker = yf.Ticker(symbol)
+
+        # Calculate start/end dynamically
+        end_date = datetime.today()
+        if period.endswith("d"):
+            days = int(period[:-1])
+        elif period.endswith("mo"):
+            days = int(period[:-2]) * 30
+        elif period.endswith("y"):
+            days = int(period[:-1]) * 365
+        else:
+            days = 30
+        start_date = end_date - timedelta(days=days)
+
+        data = ticker.history(
+            start=start_date.strftime("%Y-%m-%d"),
+            end=end_date.strftime("%Y-%m-%d")
+        )
+
+        if data.empty:
+            return f"No historical data found for '{symbol}'."
+        
+        # Convert to JSON format (reset index to include Date)
+        # return data.tail(5).reset_index().to_json(orient='records', date_format='iso')
+        return f"Historical data for {symbol} ({period}):\n{data.tail(5).to_string()}"
+
+    except Exception as e:
+        return f"Error fetching historical data for '{symbol}': {e}"
+
+def _get_analyst_recommendations(symbol: str) -> str:
+    print(f"[DEBUG] get_analyst_recommendations called for symbol='{symbol}'")
+    try:
+        ticker = yf.Ticker(symbol)
+        recs = ticker.recommendations
+        if recs is None or recs.empty:
+             return f"No analyst recommendations found for {symbol}."
+        
+        # Format the last few recommendations
+        latest = recs.tail(5)
+        return f"Analyst Recommendations for {symbol}:\n{latest.to_string()}"
+    except Exception as e:
+         return f"Error fetching recommendations for '{symbol}': {e}"
+
+def _get_earnings_calendar(symbol: str) -> str:
+    print(f"[DEBUG] get_earnings_calendar called for symbol='{symbol}'")
+    try:
+        ticker = yf.Ticker(symbol)
+        calendar = ticker.calendar
+        if calendar is None:
+            return f"No earnings calendar found for {symbol}."
+        
+        # Handle dict (new yfinance) or DataFrame (old yfinance)
+        if isinstance(calendar, dict):
+             if not calendar:
+                 return f"No earnings calendar found for {symbol}."
+        elif hasattr(calendar, 'empty') and calendar.empty:
+             return f"No earnings calendar found for {symbol}."
+            
+        return f"Earnings Calendar for {symbol}:\n{calendar}"
+    except Exception as e:
+         return f"Error fetching earnings calendar for '{symbol}': {e}"
+           
+@function_tool
+def get_summary(symbol: str, period: str = "1d", interval: str = "1h") -> str:
+    """
+    Fetch the latest summary information and intraday price data for a given ticker.
+    Ensures recent data is retrieved by calculating start/end dates dynamically.
+
+    Parameters:
+    -----------
+    symbol : str
+        The ticker symbol (e.g., "AAPL", "GOOG", "BTC-USD").
+    period : str, optional (default="1d")
+        Time range for price data. Examples: "1d", "5d", "1mo", "3mo".
+    interval : str, optional (default="1h")
+        Granularity of the data. Examples: "1m", "5m", "1h", "1d".
+
+    Returns:
+    --------
+    str
+        A formatted string containing:
+        - Company/ticker name
+        - Current price and change
+        - Open, High, Low prices
+        - Volume
+        - Period and interval used
+    """
+    return _get_summary(symbol, period, interval)
+    
+@function_tool
+def get_market_sentiment(symbol: str, period: str = "1mo") -> str:
+    """
+    Analyze recent price changes and provide a simple market sentiment.
+    Uses dynamic start/end dates to ensure recent data.
+
+    This tool computes the percentage change over the specified period and
+    classifies the sentiment as:
+    - Bullish (if price increased >2%)
+    - Bearish (if price decreased >2%)
+    - Neutral (otherwise)
+
+    Parameters:
+    -----------
+    symbol : str
+        The ticker symbol (e.g., "AAPL", "GOOG", "BTC-USD").
+    period : str, optional (default="1mo")
+        Time range to analyze. Examples: "7d", "1mo", "3mo".
+
+    Returns:
+    --------
+    str
+        A human-readable sentiment string including percentage change.
+    """
+    return _get_market_sentiment(symbol, period)
+
+@function_tool
+def get_history(symbol: str, period: str = "1mo") -> str:
+    """
+    Fetch historical price data for a given ticker.
+    Ensures recent data is retrieved dynamically using start/end dates.
+
+    Parameters:
+    -----------
+    symbol : str
+        The ticker symbol (e.g., "AAPL", "GOOG", "BTC-USD").
+    period : str, optional (default="1mo")
+        The length of historical data to retrieve. Examples: "1d", "5d", "1mo", "3mo", "1y", "5y".
+
+    Returns:
+    --------
+    str
+        A formatted string showing the last 5 rows of historical prices (Open, High, Low, Close, Volume).
+    """
+    return _get_history(symbol, period)
+
+@function_tool
+def get_analyst_recommendations(symbol: str) -> str:
+    """
+    Fetch analyst recommendations for a given ticker.
+    
+    Parameters:
+    -----------
+    symbol : str
+        The ticker symbol.
+
+    Returns:
+    --------
+    str
+        Formatted string string of analyst recommendations.
+    """
+    return _get_analyst_recommendations(symbol)
+
+@function_tool
+def get_earnings_calendar(symbol: str) -> str:
+    """
+    Fetch the next earnings date for a ticker.
+
+    Parameters:
+    -----------
+    symbol : str
+        The ticker symbol.
+
+    Returns:
+    --------
+    str
+        Next earnings date info.
+    """
+    return _get_earnings_calendar(symbol)
+

src/stock-investment-analyst/utility/autogen_model_factory.py

@@ -0,0 +1,112 @@
+import os
+from azure.identity import DefaultAzureCredential, get_bearer_token_provider
+
+class AutoGenModelFactory:
+    """
+    Factory for creating AutoGen compatible model instances.
+    """
+
+    @staticmethod
+    def get_model(provider: str = "azure", # azure, openai, google, groq, ollama
+                  model_name: str = "gpt-4o", 
+                  temperature: float = 0,
+                  model_info: dict = None
+                  ):
+        """
+        Returns an AutoGen OpenAIChatCompletionClient instance.
+        """
+        
+        # Lazy import to avoid dependency issues if autogen is not installed
+        try:
+            from autogen_ext.models.openai import OpenAIChatCompletionClient
+        except ImportError as e:
+            raise ImportError("AutoGen libraries (autogen-agentchat, autogen-ext[openai]) are not installed.") from e
+
+        # ----------------------------------------------------------------------
+        # AZURE
+        # ----------------------------------------------------------------------
+        if provider.lower() == "azure":
+            token_provider = get_bearer_token_provider(
+                DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
+            )
+            return OpenAIChatCompletionClient(
+                model=model_name,
+                azure_endpoint=os.environ["AZURE_OPENAI_API_URI"],
+                api_version=os.environ["AZURE_OPENAI_API_VERSION"],
+                azure_ad_token_provider=token_provider,
+                temperature=temperature,
+            )
+
+        # ----------------------------------------------------------------------
+        # OPENAI
+        # ----------------------------------------------------------------------
+        elif provider.lower() == "openai":
+            return OpenAIChatCompletionClient(
+                model=model_name,
+                api_key=os.environ["OPENAI_API_KEY"],
+                temperature=temperature,
+            )
+        
+        # ----------------------------------------------------------------------
+        # GOOGLE (GEMINI) via OpenAI Compat
+        # ----------------------------------------------------------------------
+        elif provider.lower() == "google" or provider.lower() == "gemini":
+            if model_info is None:
+                model_info = {
+                    "family": "gemini",
+                    "vision": False,
+                    "function_calling": True,
+                    "json_output": True,
+                    "structured_output": False
+                }
+            
+            return OpenAIChatCompletionClient(
+                model=model_name,
+                base_url="https://generativelanguage.googleapis.com/v1beta/openai/",
+                api_key=os.environ["GOOGLE_API_KEY"],
+                model_info=model_info,
+                temperature=temperature,
+                max_tokens=2048,
+                structured_output=False, # Disable for Gemini compatibility
+                extra_headers={"x-goog-api-key": os.environ["GOOGLE_API_KEY"]}
+            )
+
+        # ----------------------------------------------------------------------
+        # GROQ
+        # ----------------------------------------------------------------------
+        elif provider.lower() == "groq":
+            if model_info is None:
+                model_info = {
+                    "family": "llama", # Use llama family for Groq
+                    "vision": False,
+                    "function_calling": True,
+                    "json_output": True,
+                    "structured_output": False
+                }
+            
+            return OpenAIChatCompletionClient(
+                model=model_name,
+                base_url="https://api.groq.com/openai/v1",
+                api_key=os.environ["GROQ_API_KEY"],
+                model_info=model_info,
+                temperature=temperature,
+                structured_output=False, # Disable for Groq compatibility
+                max_tokens=2048
+            )
+        
+        # ----------------------------------------------------------------------
+        # OLLAMA
+        # ----------------------------------------------------------------------
+        elif provider.lower() == "ollama":
+            # Ensure model_info defaults to empty dict if None
+            info = model_info if model_info is not None else {}
+            return OpenAIChatCompletionClient(
+                model=model_name,
+                base_url="http://localhost:11434/v1",
+                api_key="ollama", # dummy key
+                model_info=info,
+                temperature=temperature,
+            )
+        
+        else:
+            raise ValueError(f"Unsupported AutoGen provider: {provider}")