Initial commit

.claude/commands/rag-metrics.md

@@ -0,0 +1,129 @@
+---
+description: Display system metrics and performance dashboard
+---
+
+# Metrics Dashboard Command
+
+Display comprehensive system metrics and performance statistics.
+
+## Usage
+
+No arguments required.
+
+## Instructions
+
+Shows system health, query performance, cache stats, and recent activity.
+
+## Implementation
+
+```python
+import httpx
+import json
+
+try:
+    # Get all metrics
+    response = httpx.get(
+        "http://127.0.0.1:8765/metrics",
+        timeout=10.0
+    )
+
+    if response.status_code == 200:
+        metrics = response.json()
+
+        print("=" * 70)
+        print("[=] RAG SYSTEM METRICS DASHBOARD".center(70))
+        print("=" * 70)
+        print()
+
+        # Health Status
+        health = metrics.get("health", {})
+        if health:
+            status = health.get("status", "unknown").upper()
+            status_emoji = "[OK]" if status == "HEALTHY" else "[!]"
+
+            print(f"{status_emoji} SYSTEM HEALTH: {status}")
+
+            error_rate = health.get("error_rate", 0)
+            memory_mb = health.get("memory_usage_mb", 0)
+            latency = health.get("avg_query_latency_ms", 0)
+
+            print(f"   Error Rate: {error_rate:.2%}")
+            print(f"   Memory: {memory_mb:.1f} MB")
+            print(f"   Avg Latency: {latency:.1f} ms")
+            print()
+
+        # Query Performance
+        perf = metrics.get("performance", {})
+        if perf:
+            print("[!] QUERY PERFORMANCE (7 days)")
+
+            total = perf.get("total_queries", 0)
+            qpd = perf.get("queries_per_day", 0)
+            avg_time = perf.get("avg_execution_time_ms", 0)
+            p95_time = perf.get("p95_execution_time_ms", 0)
+
+            print(f"   Total Queries: {total}")
+            print(f"   Queries/Day: {qpd:.1f}")
+            print(f"   Avg Time: {avg_time:.1f} ms")
+            print(f"   P95 Time: {p95_time:.1f} ms")
+
+            feedback = perf.get("avg_feedback_score")
+            if feedback is not None:
+                print(f"   Avg Feedback: {feedback:.2f}/1.0")
+
+            print()
+
+        # Cache Statistics
+        cache = metrics.get("cache", {})
+        if cache:
+            print("[@] CACHE STATISTICS")
+
+            entries = cache.get("total_entries", 0)
+            size_mb = cache.get("total_size_mb", 0)
+            avg_access = cache.get("average_access_count", 0)
+
+            print(f"   Total Entries: {entries}")
+            print(f"   Total Size: {size_mb:.2f} MB")
+            print(f"   Avg Access Count: {avg_access:.1f}")
+            print()
+
+        # Get popular queries
+        try:
+            history_response = httpx.get(
+                "http://127.0.0.1:8765/query-history?days=1",
+                timeout=5.0
+            )
+
+            if history_response.status_code == 200:
+                history = history_response.json()
+                popular = history.get("popular_queries", [])
+
+                if popular:
+                    print("[FIRE] TOP QUERIES (24h)")
+                    for i, query_info in enumerate(popular[:5], 1):
+                        query = query_info.get("query", "")
+                        count = query_info.get("count", 0)
+
+                        if len(query) > 50:
+                            query = query[:47] + "..."
+
+                        print(f"   {i}. {query} ({count}x)")
+
+                    print()
+
+        except:
+            pass
+
+        print("=" * 70)
+
+        print("\n[i] Refresh with: /rag-metrics")
+
+    else:
+        print(f"[X] Server error: {response.status_code}")
+
+except Exception as e:
+    print(f"[X] Error: {e}")
+    print("Make sure the MCP server is running on port 8000.")
+```
+
+Execute this code to display the metrics dashboard.