1 files changed, 96 insertions, 0 deletions
diff --git a/backend/app/services/data_service.py b/backend/app/services/data_service.py
index 9662227..108d816 100644
--- a/backend/app/services/data_service.py
+++ b/backend/app/services/data_service.py
@@ -32,6 +32,7 @@ VALUATION_CACHE = TTLCache(maxsize=128, ttl=3600)
 HIST_RATIOS_CACHE: TTLCache = TTLCache(maxsize=128, ttl=3600)
 RATIOS_ENDPOINT_CACHE: TTLCache = TTLCache(maxsize=128, ttl=3600)
 SECTOR_BENCHMARK_CACHE: TTLCache = TTLCache(maxsize=128, ttl=3600)
+INSIDERS_CACHE: TTLCache = TTLCache(maxsize=256, ttl=3600)
 
 PERIODS = {"1m", "3m", "6m", "1y", "2y", "5y"}
 YF_PERIOD_MAP = {"1m": "1mo", "3m": "3mo", "6m": "6mo", "1y": "1y", "2y": "2y", "5y": "5y"}
@@ -1770,3 +1771,98 @@ def get_ticker_overview(symbol: str) -> dict[str, Any] | None:
             "sources": field_sources,
         },
     }
+
+
+def _classify_insider(text: str) -> str:
+    t = str(text or "").lower()
+    if any(k in t for k in ("sale", "sold", "disposition")):
+        return "sell"
+    if any(k in t for k in ("purchase", "bought", "acquisition", "grant", "award", "exercise")):
+        return "buy"
+    return "other"
+
+
+@cached(INSIDERS_CACHE)
+def get_insider_transactions(symbol: str) -> dict:
+    sym = normalize_symbol(symbol)
+    try:
+        t = yf.Ticker(sym)
+        df = t.insider_transactions
+    except Exception:
+        df = None
+
+    if df is None or (hasattr(df, "empty") and df.empty):
+        return {
+            "summary": {"buy_count": 0, "sell_count": 0, "buy_value": 0.0, "sell_value": 0.0},
+            "monthly_chart": [],
+            "transactions": [],
+        }
+
+    df = df.copy()
+    df["direction"] = df["Text"].apply(_classify_insider)
+
+    def _to_dt(val: object) -> "pd.Timestamp | None":
+        try:
+            return pd.to_datetime(val)
+        except Exception:
+            return None
+
+    df["_date"] = df["Start Date"].apply(_to_dt)
+
+    cutoff = pd.Timestamp.now() - pd.Timedelta(days=180)
+    recent = df[df["_date"] >= cutoff]
+
+    def _total_value(subset: "pd.DataFrame") -> float:
+        try:
+            return float(subset["Value"].dropna().astype(float).sum())
+        except Exception:
+            return 0.0
+
+    buys = recent[recent["direction"] == "buy"]
+    sells = recent[recent["direction"] == "sell"]
+
+    summary = {
+        "buy_count": int(len(buys)),
+        "sell_count": int(len(sells)),
+        "buy_value": _total_value(buys),
+        "sell_value": _total_value(sells),
+    }
+
+    # Monthly chart: last 6 months, buys positive, sells negative (in $M)
+    monthly: dict[str, dict[str, float]] = {}
+    for _, row in recent.iterrows():
+        dt = row["_date"]
+        if dt is None or pd.isna(dt):
+            continue
+        key = dt.strftime("%Y-%m")
+        monthly.setdefault(key, {"buy": 0.0, "sell": 0.0})
+        try:
+            val = float(row["Value"]) if pd.notna(row["Value"]) else 0.0
+        except (TypeError, ValueError):
+            val = 0.0
+        if row["direction"] in ("buy", "sell"):
+            monthly[key][row["direction"]] += val
+
+    monthly_chart = [
+        {"month": m, "buy": monthly[m]["buy"] / 1e6, "sell": monthly[m]["sell"] / 1e6}
+        for m in sorted(monthly.keys())
+    ]
+
+    # All transactions (newest first)
+    transactions = []
+    for _, row in df.sort_values("_date", ascending=False).iterrows():
+        dt = row["_date"]
+        transactions.append({
+            "date": dt.strftime("%Y-%m-%d") if dt is not None and not pd.isna(dt) else None,
+            "insider": str(row.get("Insider", "") or ""),
+            "position": str(row.get("Position", "") or ""),
+            "direction": row["direction"],
+            "shares": int(row["Shares"]) if pd.notna(row.get("Shares")) else None,
+            "value": _safe_float(row.get("Value")),
+        })
+
+    return {
+        "summary": summary,
+        "monthly_chart": monthly_chart,
+        "transactions": transactions,
+    }