Add Yahoo Finance integration as free data provider option

app/backend/models/schemas.py

@@ -65,6 +65,7 @@ class BaseHedgeFundRequest(BaseModel):
     agent_models: Optional[List[AgentModelConfig]] = None
     model_name: Optional[str] = "gpt-4.1"
     model_provider: Optional[ModelProvider] = ModelProvider.OPENAI
+    data_provider: Optional[str] = "yfinance"  # Default to free option
     margin_requirement: float = 0.0
     portfolio_positions: Optional[List[PortfolioPosition]] = None
     api_keys: Optional[Dict[str, str]] = None

app/backend/services/graph.py

@@ -129,12 +129,12 @@ def create_graph(graph_nodes: list, graph_edges: list) -> StateGraph:
     return graph
 
 
-async def run_graph_async(graph, portfolio, tickers, start_date, end_date, model_name, model_provider, request=None):
+async def run_graph_async(graph, portfolio, tickers, start_date, end_date, model_name, model_provider, data_provider="yfinance", request=None):
     """Async wrapper for run_graph to work with asyncio."""
     # Use run_in_executor to run the synchronous function in a separate thread
     # so it doesn't block the event loop
     loop = asyncio.get_running_loop()
-    result = await loop.run_in_executor(None, lambda: run_graph(graph, portfolio, tickers, start_date, end_date, model_name, model_provider, request))  # Use default executor
+    result = await loop.run_in_executor(None, lambda: run_graph(graph, portfolio, tickers, start_date, end_date, model_name, model_provider, data_provider, request))  # Use default executor
     return result
 
 
@@ -146,6 +146,7 @@ def run_graph(
     end_date: str,
     model_name: str,
     model_provider: str,
+    data_provider: str = "yfinance",
     request=None,
 ) -> dict:
     """
@@ -171,6 +172,7 @@ def run_graph(
                 "show_reasoning": False,
                 "model_name": model_name,
                 "model_provider": model_provider,
+                "data_provider": data_provider,
                 "request": request,  # Pass the request for agent-specific model access
             },
         },

pyproject.toml

@@ -35,6 +35,7 @@ httpx = "^0.27.0"
 sqlalchemy = "^2.0.22"
 alembic = "^1.12.0"
 langchain-gigachat = "^0.3.12"
+yfinance = "^0.2.65"
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^7.4.0"

src/agents/aswath_damodaran.py

@@ -13,6 +13,7 @@
     get_market_cap,
     search_line_items,
 )
+from src.data.providers import get_data_provider_for_agent
 from src.utils.api_key import get_api_key_from_state
 from src.utils.llm import call_llm
 from src.utils.progress import progress
@@ -37,14 +38,16 @@ def aswath_damodaran_agent(state: AgentState, agent_id: str = "aswath_damodaran_
     end_date  = data["end_date"]
     tickers   = data["tickers"]
     api_key  = get_api_key_from_state(state, "FINANCIAL_DATASETS_API_KEY")
+    # Use centralized data provider configuration
+    data_provider = get_data_provider_for_agent(state, agent_id)
 
     analysis_data: dict[str, dict] = {}
     damodaran_signals: dict[str, dict] = {}
 
     for ticker in tickers:
         # ─── Fetch core data ────────────────────────────────────────────────────
         progress.update_status(agent_id, ticker, "Fetching financial metrics")
-        metrics = get_financial_metrics(ticker, end_date, period="ttm", limit=5, api_key=api_key)
+        metrics = get_financial_metrics(ticker, end_date, period="ttm", limit=5, api_key=api_key, data_provider=data_provider)
 
         progress.update_status(agent_id, ticker, "Fetching financial line items")
         line_items = search_line_items(
@@ -61,10 +64,11 @@ def aswath_damodaran_agent(state: AgentState, agent_id: str = "aswath_damodaran_
             ],
             end_date,
             api_key=api_key,
+            data_provider=data_provider,
         )
 
         progress.update_status(agent_id, ticker, "Getting market cap")
-        market_cap = get_market_cap(ticker, end_date, api_key=api_key)
+        market_cap = get_market_cap(ticker, end_date, api_key=api_key, data_provider=data_provider)
 
         # ─── Analyses ───────────────────────────────────────────────────────────
         progress.update_status(agent_id, ticker, "Analyzing growth and reinvestment")

src/agents/ben_graham.py

@@ -1,5 +1,6 @@
 from src.graph.state import AgentState, show_agent_reasoning
 from src.tools.api import get_financial_metrics, get_market_cap, search_line_items
+from src.data.providers import get_data_provider_for_agent
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.messages import HumanMessage
 from pydantic import BaseModel
@@ -29,19 +30,21 @@ def ben_graham_agent(state: AgentState, agent_id: str = "ben_graham_agent"):
     end_date = data["end_date"]
     tickers = data["tickers"]
     api_key = get_api_key_from_state(state, "FINANCIAL_DATASETS_API_KEY")
+    # Use centralized data provider configuration
+    data_provider = get_data_provider_for_agent(state, agent_id)
 
     analysis_data = {}
     graham_analysis = {}
 
     for ticker in tickers:
         progress.update_status(agent_id, ticker, "Fetching financial metrics")
-        metrics = get_financial_metrics(ticker, end_date, period="annual", limit=10, api_key=api_key)
+        metrics = get_financial_metrics(ticker, end_date, period="annual", limit=10, api_key=api_key, data_provider=data_provider)
 
         progress.update_status(agent_id, ticker, "Gathering financial line items")
-        financial_line_items = search_line_items(ticker, ["earnings_per_share", "revenue", "net_income", "book_value_per_share", "total_assets", "total_liabilities", "current_assets", "current_liabilities", "dividends_and_other_cash_distributions", "outstanding_shares"], end_date, period="annual", limit=10, api_key=api_key)
+        financial_line_items = search_line_items(ticker, ["earnings_per_share", "revenue", "net_income", "book_value_per_share", "total_assets", "total_liabilities", "current_assets", "current_liabilities", "dividends_and_other_cash_distributions", "outstanding_shares"], end_date, period="annual", limit=10, api_key=api_key, data_provider=data_provider)
 
         progress.update_status(agent_id, ticker, "Getting market cap")
-        market_cap = get_market_cap(ticker, end_date, api_key=api_key)
+        market_cap = get_market_cap(ticker, end_date, api_key=api_key, data_provider=data_provider)
 
         # Perform sub-analyses
         progress.update_status(agent_id, ticker, "Analyzing earnings stability")
@@ -109,7 +112,7 @@ def analyze_earnings_stability(metrics: list, financial_line_items: list) -> dic
 
     eps_vals = []
     for item in financial_line_items:
-        if item.earnings_per_share is not None:
+        if hasattr(item, "earnings_per_share") and item.earnings_per_share is not None:
             eps_vals.append(item.earnings_per_share)
 
     if len(eps_vals) < 2:
@@ -149,11 +152,10 @@ def analyze_financial_strength(financial_line_items: list) -> dict:
     if not financial_line_items:
         return {"score": score, "details": "No data for financial strength analysis"}
 
-    latest_item = financial_line_items[0]
-    total_assets = latest_item.total_assets or 0
-    total_liabilities = latest_item.total_liabilities or 0
-    current_assets = latest_item.current_assets or 0
-    current_liabilities = latest_item.current_liabilities or 0
+    total_assets = get_line_item_value(financial_line_items, "total_assets")
+    total_liabilities = get_line_item_value(financial_line_items, "total_liabilities")
+    current_assets = get_line_item_value(financial_line_items, "current_assets")
+    current_liabilities = get_line_item_value(financial_line_items, "current_liabilities")
 
     # 1. Current ratio
     if current_liabilities > 0:
@@ -184,7 +186,7 @@ def analyze_financial_strength(financial_line_items: list) -> dict:
         details.append("Cannot compute debt ratio (missing total_assets).")
 
     # 3. Dividend track record
-    div_periods = [item.dividends_and_other_cash_distributions for item in financial_line_items if item.dividends_and_other_cash_distributions is not None]
+    div_periods = [item.dividends_and_other_cash_distributions for item in financial_line_items if hasattr(item, "dividends_and_other_cash_distributions") and item.dividends_and_other_cash_distributions is not None]
     if div_periods:
         # In many data feeds, dividend outflow is shown as a negative number
         # (money going out to shareholders). We'll consider any negative as 'paid a dividend'.
@@ -204,6 +206,15 @@ def analyze_financial_strength(financial_line_items: list) -> dict:
     return {"score": score, "details": "; ".join(details)}
 
 
+def get_line_item_value(financial_line_items: list, line_item_name: str) -> float:
+    """Helper function to extract value for a specific line item."""
+    for item in financial_line_items:
+        if hasattr(item, line_item_name):
+            value = getattr(item, line_item_name)
+            if value is not None:
+                return value
+    return 0
+
 def analyze_valuation_graham(financial_line_items: list, market_cap: float) -> dict:
     """
     Core Graham approach to valuation:
@@ -214,12 +225,11 @@ def analyze_valuation_graham(financial_line_items: list, market_cap: float) -> d
     if not financial_line_items or not market_cap or market_cap <= 0:
         return {"score": 0, "details": "Insufficient data to perform valuation"}
 
-    latest = financial_line_items[0]
-    current_assets = latest.current_assets or 0
-    total_liabilities = latest.total_liabilities or 0
-    book_value_ps = latest.book_value_per_share or 0
-    eps = latest.earnings_per_share or 0
-    shares_outstanding = latest.outstanding_shares or 0
+    current_assets = get_line_item_value(financial_line_items, "current_assets")
+    total_liabilities = get_line_item_value(financial_line_items, "total_liabilities")
+    book_value_ps = get_line_item_value(financial_line_items, "book_value_per_share")
+    eps = get_line_item_value(financial_line_items, "earnings_per_share")
+    shares_outstanding = get_line_item_value(financial_line_items, "outstanding_shares")
 
     details = []
     score = 0

src/agents/bill_ackman.py

@@ -1,5 +1,6 @@
 from src.graph.state import AgentState, show_agent_reasoning
 from src.tools.api import get_financial_metrics, get_market_cap, search_line_items
+from src.data.providers import get_data_provider_for_agent
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.messages import HumanMessage
 from pydantic import BaseModel
@@ -26,12 +27,14 @@ def bill_ackman_agent(state: AgentState, agent_id: str = "bill_ackman_agent"):
     end_date = data["end_date"]
     tickers = data["tickers"]
     api_key = get_api_key_from_state(state, "FINANCIAL_DATASETS_API_KEY")
+    # Use centralized data provider configuration
+    data_provider = get_data_provider_for_agent(state, agent_id)
     analysis_data = {}
     ackman_analysis = {}
 
     for ticker in tickers:
         progress.update_status(agent_id, ticker, "Fetching financial metrics")
-        metrics = get_financial_metrics(ticker, end_date, period="annual", limit=5, api_key=api_key)
+        metrics = get_financial_metrics(ticker, end_date, period="annual", limit=5, api_key=api_key, data_provider=data_provider)
 
         progress.update_status(agent_id, ticker, "Gathering financial line items")
         # Request multiple periods of data (annual or TTM) for a more robust long-term view.
@@ -53,10 +56,11 @@ def bill_ackman_agent(state: AgentState, agent_id: str = "bill_ackman_agent"):
             period="annual",
             limit=5,
             api_key=api_key,
+            data_provider=data_provider,
         )
 
         progress.update_status(agent_id, ticker, "Getting market cap")
-        market_cap = get_market_cap(ticker, end_date, api_key=api_key)
+        market_cap = get_market_cap(ticker, end_date, api_key=api_key, data_provider=data_provider)
 
         progress.update_status(agent_id, ticker, "Analyzing business quality")
         quality_analysis = analyze_business_quality(metrics, financial_line_items)
@@ -150,7 +154,7 @@ def analyze_business_quality(metrics: list, financial_line_items: list) -> dict:
         }
 
     # 1. Multi-period revenue growth analysis
-    revenues = [item.revenue for item in financial_line_items if item.revenue is not None]
+    revenues = [getattr(item, 'revenue', None) for item in financial_line_items if getattr(item, 'revenue', None) is not None]
     if len(revenues) >= 2:
         initial, final = revenues[-1], revenues[0]
         if initial and final and final > initial:
@@ -167,8 +171,8 @@ def analyze_business_quality(metrics: list, financial_line_items: list) -> dict:
         details.append("Not enough revenue data for multi-period trend.")
 
     # 2. Operating margin and free cash flow consistency
-    fcf_vals = [item.free_cash_flow for item in financial_line_items if item.free_cash_flow is not None]
-    op_margin_vals = [item.operating_margin for item in financial_line_items if item.operating_margin is not None]
+    fcf_vals = [getattr(item, 'free_cash_flow', None) for item in financial_line_items if getattr(item, 'free_cash_flow', None) is not None]
+    op_margin_vals = [getattr(item, 'operating_margin', None) for item in financial_line_items if getattr(item, 'operating_margin', None) is not None]
 
     if op_margin_vals:
         above_15 = sum(1 for m in op_margin_vals if m > 0.15)
@@ -228,7 +232,7 @@ def analyze_financial_discipline(metrics: list, financial_line_items: list) -> d
         }
 
     # 1. Multi-period debt ratio or debt_to_equity
-    debt_to_equity_vals = [item.debt_to_equity for item in financial_line_items if item.debt_to_equity is not None]
+    debt_to_equity_vals = [getattr(item, 'debt_to_equity', None) for item in financial_line_items if getattr(item, 'debt_to_equity', None) is not None]
     if debt_to_equity_vals:
         below_one_count = sum(1 for d in debt_to_equity_vals if d < 1.0)
         if below_one_count >= (len(debt_to_equity_vals) // 2 + 1):
@@ -255,9 +259,9 @@ def analyze_financial_discipline(metrics: list, financial_line_items: list) -> d
 
     # 2. Capital allocation approach (dividends + share counts)
     dividends_list = [
-        item.dividends_and_other_cash_distributions
+        getattr(item, 'dividends_and_other_cash_distributions', None)
         for item in financial_line_items
-        if item.dividends_and_other_cash_distributions is not None
+        if getattr(item, 'dividends_and_other_cash_distributions', None) is not None
     ]
     if dividends_list:
         paying_dividends_count = sum(1 for d in dividends_list if d < 0)
@@ -270,7 +274,7 @@ def analyze_financial_discipline(metrics: list, financial_line_items: list) -> d
         details.append("No dividend data found across periods.")
 
     # Check for decreasing share count (simple approach)
-    shares = [item.outstanding_shares for item in financial_line_items if item.outstanding_shares is not None]
+    shares = [getattr(item, 'outstanding_shares', None) for item in financial_line_items if getattr(item, 'outstanding_shares', None) is not None]
     if len(shares) >= 2:
         # For buybacks, the newest count should be less than the oldest count
         if shares[0] < shares[-1]:
@@ -303,8 +307,8 @@ def analyze_activism_potential(financial_line_items: list) -> dict:
         }
 
     # Check revenue growth vs. operating margin
-    revenues = [item.revenue for item in financial_line_items if item.revenue is not None]
-    op_margins = [item.operating_margin for item in financial_line_items if item.operating_margin is not None]
+    revenues = [getattr(item, 'revenue', None) for item in financial_line_items if getattr(item, 'revenue', None) is not None]
+    op_margins = [getattr(item, 'operating_margin', None) for item in financial_line_items if getattr(item, 'operating_margin', None) is not None]
 
     if len(revenues) < 2 or not op_margins:
         return {