P5-PR14: polish — admin health alias at /admin/health-status; rate limit headers add X-RateLimit-Reset; keep single limiter mount

Author: DMontgomery40

api/app/main.py

@@ -905,6 +905,38 @@ def health_ready():
     )
 
 
+@app.get("/admin/health-status", dependencies=[Depends(require_admin)])
+async def admin_health_status():
+    """Return provider health summary (alias for /admin/providers/health)."""
+    try:
+        from .routers.admin_providers import get_provider_health_status  # type: ignore
+        from fastapi import Request as _Req
+        # Fabricate a minimal Request-like object with app reference
+        class _R:
+            def __init__(self, app):
+                self.app = app
+        return await get_provider_health_status(_R(app))  # type: ignore
+    except Exception:
+        hm = getattr(app.state, "health_monitor", None)
+        if not hm:
+            raise HTTPException(503, detail="Health monitor not available")
+        statuses = await hm.get_provider_statuses()
+        # Summarize
+        counts = {k: 0 for k in ["healthy", "degraded", "circuit_open", "disabled"]}
+        for s in statuses.values():
+            st = s.get("status", "healthy")
+            if st in counts:
+                counts[st] += 1
+        return {
+            "provider_statuses": statuses,
+            "total_providers": len(statuses),
+            "healthy_count": counts["healthy"],
+            "degraded_count": counts["degraded"],
+            "circuit_open_count": counts["circuit_open"],
+            "disabled_count": counts["disabled"],
+        }
+
+
 def require_api_key(request: Request, x_api_key: Optional[str] = Header(default=None)):
     """Authenticate request using either env API_KEY or DB-backed key.
     Behavior:

api/app/middleware/hierarchical_rate_limiter.py

@@ -212,10 +212,18 @@ def _rate_limit_response(self, limit: int) -> Response:
             "retry_after_seconds": retry_after
         }
 
+        # Compute seconds until window reset; fall back to retry_after on error
+        try:
+            now = int(time.time())
+            seconds = 60 - (now % 60)
+        except Exception:
+            seconds = retry_after
+
         headers = {
             "Retry-After": str(retry_after),
             "X-RateLimit-Limit": str(limit),
             "X-RateLimit-Remaining": "0",
+            "X-RateLimit-Reset": str(seconds),
         }
 
         return Response(
@@ -224,3 +232,12 @@ def _rate_limit_response(self, limit: int) -> Response:
             headers=headers,
             media_type="application/json"
         )
+
+    def remaining_for(self, bucket_key: str, limit: int) -> int:
+        try:
+            bucket = self.rate_buckets.get(bucket_key, {"count": 0})
+            used = int(bucket.get("count", 0))
+            remaining = max(0, limit - used)
+            return remaining
+        except Exception:
+            return 0