Restrict observability endpoints to ops admins

frontend/app/api/system/status/route.ts

@@ -1,5 +1,10 @@
 import { NextRequest, NextResponse } from "next/server";
-import { proxyBackendJsonGet } from "@/lib/api-proxy";
+import { buildProxyExceptionResponse } from "@/lib/api-proxy";
+import {
+  applyAuthResponseCookies,
+  buildBackendRequestHeaders,
+} from "@/lib/backend-auth";
+import { requireOpsProxyAuth } from "@/lib/ops-proxy-auth";
 
 const API_BASE = process.env.POLYWEATHER_API_BASE_URL;
 
@@ -11,11 +16,28 @@ export async function GET(req: NextRequest) {
     );
   }
 
-  return proxyBackendJsonGet(req, {
+  try {
-    cacheControl: "public, max-age=0, s-maxage=30, stale-while-revalidate=120",
+    const auth = await buildBackendRequestHeaders(req);
-    detailLimit: 500,
+    const authError = requireOpsProxyAuth(req, auth);
-    publicMessage: "Failed to fetch system status",
+    if (authError) return authError;
-    revalidateSeconds: 30,
+
-    url: `${API_BASE}/api/system/status`,
+    const res = await fetch(`${API_BASE}/api/system/status`, {
-  });
+      cache: "no-store",
+      headers: auth.headers,
+    });
+    const raw = await res.text();
+    const response = new NextResponse(raw, {
+      headers: {
+        "Cache-Control": "no-store",
+        "Cloudflare-CDN-Cache-Control": "no-store",
+        "Content-Type": res.headers.get("content-type") || "application/json",
+      },
+      status: res.status,
+    });
+    return applyAuthResponseCookies(response, auth.response);
+  } catch (error) {
+    return buildProxyExceptionResponse(error, {
+      publicMessage: "Failed to fetch system status",
+    });
+  }
 }

frontend/components/account/__tests__/paymentSecurity.test.ts

@@ -186,6 +186,7 @@ export function runTests() {
     "app/api/ops/users/route.ts",
     "app/api/ops/users/grant-points/route.ts",
     "app/api/ops/view-logs/route.ts",
+    "app/api/system/status/route.ts",
   ]) {
     const routeSource = fs.readFileSync(path.join(projectRoot, route), "utf8");
     assert(
@@ -211,14 +212,10 @@ export function runTests() {
   const optionalRefreshIndex = middlewareSource.indexOf(
     "function shouldRefreshOptionalSupabaseSession",
   );
-  const systemStatusPublicIndex = middlewareSource.indexOf(
-    'pathname === "/api/system/status"',
-  );
   assert(
-    systemStatusPublicIndex >= 0 &&
+    optionalRefreshIndex >= 0 &&
-      optionalRefreshIndex >= 0 &&
+      !middlewareSource.includes('pathname === "/api/system/status"'),
-      systemStatusPublicIndex < optionalRefreshIndex,
+    "middleware must not treat system status as public after it becomes an ops-only API",
-    "middleware must treat public system status as public API instead of optional Supabase session refresh",
   );
 
   for (const route of paymentRoutes) {

frontend/middleware.ts

@@ -36,7 +36,6 @@ function isPublicApi(pathname: string) {
     pathname === "/api/cities" ||
     pathname === "/api/payments/config" ||
     pathname === "/api/scan/terminal" ||
-    pathname === "/api/system/status" ||
     pathname === "/api/vitals" ||
     /^\/api\/city\/[^/]+$/i.test(pathname) ||
     /^\/api\/city\/[^/]+\/summary$/i.test(pathname) ||

tests/test_web_observability.py

@@ -52,7 +52,18 @@ def test_healthz_keeps_liveness_200_when_db_health_is_degraded(monkeypatch):
     assert response.json()["status"] == "degraded"
 
 
-def test_system_status_returns_summary_shape():
+def test_system_status_requires_ops_admin():
+    response = client.get('/api/system/status')
+    assert response.status_code in {401, 403, 503}
+
+
+def test_system_status_returns_summary_shape_for_ops_admin(monkeypatch):
+    monkeypatch.setattr(
+        routes,
+        "_require_ops_admin",
+        lambda request: {"user_id": "admin-user", "email": "admin@example.com"},
+    )
+
     response = client.get('/api/system/status')
     assert response.status_code == 200
     payload = response.json()
@@ -93,12 +104,31 @@ def test_observation_freshness_accepts_epoch_seconds():
     assert payload["observed_at"].startswith("2026-")
 
 
-def test_metrics_endpoint_returns_prometheus_payload():
+def test_metrics_endpoint_requires_ops_admin():
+    response = client.get('/metrics')
+    assert response.status_code in {401, 403, 503}
+
+
+def test_metrics_endpoint_returns_prometheus_payload_for_ops_admin(monkeypatch):
+    monkeypatch.setattr(
+        routes,
+        "_require_ops_admin",
+        lambda request: {"user_id": "admin-user", "email": "admin@example.com"},
+    )
+
     response = client.get('/metrics')
     assert response.status_code == 200
     assert 'polyweather_http_requests_total' in response.text
 
 
+def test_system_cache_status_requires_ops_admin(monkeypatch):
+    monkeypatch.setattr(routes, "_assert_entitlement", lambda request: None)
+
+    response = client.get("/api/system/cache-status?cities=shanghai")
+
+    assert response.status_code in {401, 403, 503}
+
+
 def test_standard_growth_funnel_events_are_trackable():
     assert {
         "landing_view",

web/routers/system.py

@@ -23,8 +23,8 @@ async def healthz():
 
 
 @router.get("/api/system/status")
-async def system_status():
+async def system_status(request: Request):
-    return await get_system_status_payload()
+    return await get_system_status_payload(request)
 
 
 @router.get("/api/system/cache-status")
@@ -42,8 +42,8 @@ async def system_priority_warm(
 
 
 @router.get("/metrics", response_class=PlainTextResponse)
-async def metrics():
+async def metrics(request: Request):
-    return get_prometheus_metrics_response()
+    return get_prometheus_metrics_response(request)
 
 
 @router.get("/api/dashboard/init")

web/services/system_api.py

@@ -19,7 +19,8 @@ def get_health_payload() -> Dict[str, Any]:
     return build_health_payload()
 
 
-async def get_system_status_payload() -> Dict[str, Any]:
+async def get_system_status_payload(request: Request) -> Dict[str, Any]:
+    legacy_routes._require_ops_admin(request)
     payload = await run_in_threadpool(build_system_status_payload)
     payload["realtime"] = await run_in_threadpool(_realtime_status_payload)
     return payload
@@ -52,7 +53,7 @@ def _realtime_status_payload() -> Dict[str, Any]:
 
 
 def get_system_cache_status(request: Request, cities: Optional[str] = None) -> Dict[str, Any]:
-    legacy_routes._assert_entitlement(request)
+    legacy_routes._require_ops_admin(request)
     selected = legacy_routes._normalize_city_list(cities)
     if not selected:
         selected = legacy_routes._normalize_city_list(None)
@@ -118,7 +119,8 @@ def run_system_priority_warm(
     }
 
 
-def get_prometheus_metrics_response() -> PlainTextResponse:
+def get_prometheus_metrics_response(request: Request) -> PlainTextResponse:
+    legacy_routes._require_ops_admin(request)
     return PlainTextResponse(
         export_prometheus_metrics(),
         media_type="text/plain; version=0.0.4; charset=utf-8",