Maximize Qdrant utilization: richer memory payloads, filtered retrieval, and qdrant health in metrics API

app.py

@@ -6,7 +6,7 @@ from datetime import datetime, timedelta
 import threading
 from db import init_db, SessionLocal, BotDecision, TradeExecution, CuratedInsight
 from bot import start_scheduler, run_cycle, curate_cycle
-from services import account_snapshot
+from services import account_snapshot, qdrant_memory_health
 
 app = FastAPI(title="alpaca-llm-bot-v1")
 templates = Jinja2Templates(directory="templates")
@@ -94,6 +94,7 @@ def api_metrics(hours: int = 72):
                 "cumulativeNotional": t_values,
             },
             "leaderboard": leaderboard,
+            "qdrant": qdrant_memory_health(),
         }
     finally:
         db.close()

bot.py

@@ -125,9 +125,11 @@ def run_cycle():
 
                 # learning memory + notes + orchestration signal
                 qdrant_add_memory(symbol, f"{symbol} {decision['action']} conf={decision['confidence']} reason={decision['reason']}", {
+                    "memory_type": "execution",
                     "status": drow.status,
                     "action": decision["action"],
                     "confidence": decision["confidence"],
+                    "outcome_score": 1 if ok else -1,
                     "ts": datetime.utcnow().isoformat(),
                 })
                 trilium_log(
@@ -145,9 +147,11 @@ def run_cycle():
 
             # store non-trade decisions too for memory
             qdrant_add_memory(symbol, f"{symbol} decision={decision['action']} conf={decision['confidence']} status={drow.status}", {
+                "memory_type": "decision",
                 "status": drow.status,
                 "action": decision["action"],
                 "confidence": decision["confidence"],
+                "outcome_score": 0 if drow.status in {"skipped", "risk_blocked"} else (-1 if drow.status=="failed" else 1),
                 "ts": datetime.utcnow().isoformat(),
             })
 

services.py

@@ -34,8 +34,15 @@ def qdrant_add_memory(symbol: str, text: str, payload: dict):
     if not vec:
         return False
     qdrant_ensure_collection(len(vec))
+    enriched = {
+        "symbol": symbol,
+        "text": text[:2000],
+        "memory_type": payload.get("memory_type", "decision"),
+        "created_at": int(time.time()),
+        **payload,
+    }
     body = {
-        "points": [{"id": str(uuid.uuid4()), "vector": vec, "payload": {"symbol": symbol, "text": text[:2000], **payload}}]
+        "points": [{"id": str(uuid.uuid4()), "vector": vec, "payload": enriched}]
     }
     try:
         r = requests.put(f"{settings.qdrant_url}/collections/{settings.qdrant_collection}/points", json=body, timeout=15)
@@ -44,11 +51,21 @@ def qdrant_add_memory(symbol: str, text: str, payload: dict):
         return False
 
 
-def qdrant_similar(symbol: str, query_text: str, limit: int = 5):
+def qdrant_similar(symbol: str, query_text: str, limit: int = 8):
     vec = _embed(query_text)
     if not vec:
         return []
-    body = {"vector": vec, "limit": limit, "with_payload": True}
+    body = {
+        "vector": vec,
+        "limit": limit,
+        "with_payload": True,
+        "filter": {
+            "should": [
+                {"key": "symbol", "match": {"value": symbol}},
+                {"key": "memory_type", "match": {"value": "macro"}}
+            ]
+        }
+    }
     try:
         r = requests.post(f"{settings.qdrant_url}/collections/{settings.qdrant_collection}/points/search", json=body, timeout=15)
         if not r.ok:
@@ -56,8 +73,14 @@ def qdrant_similar(symbol: str, query_text: str, limit: int = 5):
         out = []
         for p in r.json().get("result", []):
             pl = p.get("payload", {})
-            if pl.get("symbol") in (symbol, None):
-                out.append({"score": p.get("score", 0), "text": pl.get("text", ""), "status": pl.get("status", "")})
+            out.append({
+                "score": p.get("score", 0),
+                "text": pl.get("text", ""),
+                "status": pl.get("status", ""),
+                "action": pl.get("action", ""),
+                "confidence": pl.get("confidence", None),
+                "outcome_score": pl.get("outcome_score", None),
+            })
         return out
     except Exception:
         return []
@@ -233,6 +256,22 @@ def trilium_log(title: str, body: str):
     return False
 
 
+def qdrant_memory_health():
+    try:
+        r = requests.get(f"{settings.qdrant_url}/collections/{settings.qdrant_collection}", timeout=10)
+        if not r.ok:
+            return {"ok": False, "status": r.status_code}
+        j = r.json().get("result", {})
+        return {
+            "ok": True,
+            "collection": settings.qdrant_collection,
+            "points_count": j.get("points_count"),
+            "indexed_vectors_count": j.get("indexed_vectors_count"),
+        }
+    except Exception as e:
+        return {"ok": False, "error": str(e)[:160]}
+
+
 def n8n_emit(event: dict):
     if not settings.n8n_webhook:
         return False