fix(cli): use server-side deployment metrics summaries

Request summary=true and pass through the server summary instead of downloading and aggregating chart data locally. Preserve compact and raw output and existing time-window defaults.

Co-authored-by: Robin <open-swe@users.noreply.github.com>

libs/cli/README.md

@@ -129,8 +129,9 @@ The top level includes `deployment_id`, `start_time`, `end_time`, and
 point bounds. The CLI resolves omitted bounds once: end defaults to now (UTC),
 and start defaults to 15 minutes before the resolved end. With only `--end-time`,
 start is 15 minutes before that value; with only `--start-time`, end is now. Both
-values are sent explicitly to the backend and output, even when there are no points. `step_seconds` is the most common gap between consecutive
-timestamps within series, with ties broken by the smallest gap. Whole seconds
+values are sent explicitly to the backend and included in default compact output,
+even when there are no points. `step_seconds` is the most common gap between
+consecutive timestamps within series, with ties broken by the smallest gap. Whole seconds
 are integers; fractional seconds are preserved. It is `null` only when every
 series has fewer than two timestamps. UI metadata and repeated series IDs are
 omitted. The host-backend response itself is unchanged.
@@ -140,16 +141,20 @@ langgraph deploy metrics DEPLOYMENT_ID --summary
 langgraph deploy metrics DEPLOYMENT_ID --raw
 ```
 
-`--summary` returns the same shape without `points`. `--raw` returns the original
-UI chart schema without reshaping or rounding. These flags are mutually exclusive.
-Neither mode returns traces.
+`--summary` requests server-side aggregation with `summary=true` and prints the
+response without reshaping or rounding. It includes `deployment_id`,
+`step_seconds`, and `sections`, but no `start_time`, `end_time`, or `points`.
+Series contain `metric`, `unit`, `latest`, `min`, `max`, and `avg`; units include
+`percent`, `count`, `milliseconds`, and `responses_per_second`.
+`--raw` returns the original UI chart schema without reshaping or rounding.
+These flags are mutually exclusive. Neither mode returns traces.
 
 Size comparison using the realistic synthetic unit-test fixture (compact JSON
-serialization for all modes; not a live deployment benchmark):
+serialization; not a live deployment benchmark):
 
-| Before: raw UI schema | After: default | After: summary |
-| --- | --- | --- |
-| 4,302 bytes | 1,226 bytes (72% smaller) | 806 bytes (81% smaller) |
+| Before: raw UI schema | After: default |
+| --- | --- |
+| 4,302 bytes | 1,226 bytes (72% smaller) |
 
 ## Configuration
 

libs/cli/langgraph_cli/deploy.py

@@ -2694,19 +2694,20 @@ def deploy_metrics(
         }.items()
         if value is not None
     }
+    if summary:
+        params["summary"] = "true"
     client = _create_host_backend_client(host_url, api_key, no_input=True)
     result = _call_host_backend_with_optional_tenant(
         client,
         lambda c: c.get_deployment_metrics(str(deployment_id), params),
         no_input=True,
     )
-    if not raw:
+    if not raw and not summary:
         result = compact_deployment_metrics(
             result,
             str(deployment_id),
             start_time=start_time,
             end_time=end_time,
-            summary=summary,
         )
     click.echo(json_mod.dumps(result, separators=(",", ":")))
 

libs/cli/langgraph_cli/deployment_metrics.py

@@ -38,7 +38,6 @@ def compact_deployment_metrics(
     *,
     start_time: str,
     end_time: str,
-    summary: bool = False,
 ) -> dict[str, Any]:
     """Reshape UI charts without mutating the response; step is in seconds."""
     sections = {}
@@ -74,14 +73,13 @@ def compact_deployment_metrics(
                     "max": round(max(values), 3) if values else None,
                     "avg": round(fmean(values), 3) if values else None,
                 }
-                if not summary:
-                    stats["points"] = [
-                        [
-                            _iso(timestamp),
-                            round(value, 3) if value is not None else None,
-                        ]
-                        for timestamp, value in points
+                stats["points"] = [
+                    [
+                        _iso(timestamp),
+                        round(value, 3) if value is not None else None,
                     ]
+                    for timestamp, value in points
+                ]
                 series_output[series["name"]] = stats
             charts[chart["title"]] = series_output or {"_empty": True}
         sections[section["title"]] = charts or {"_empty": True}

libs/cli/tests/unit_tests/cli/test_deploy_monitoring.py

@@ -34,6 +34,30 @@ def invoke(monkeypatch, args, handler, input=None):
 @pytest.mark.parametrize("mode", [None, "--summary", "--raw"])
 def test_metrics(monkeypatch, bounds, mode, deployment_metrics_response):
     charts = deployment_metrics_response
+    server_summary = {
+        "deployment_id": DEPLOYMENT_ID,
+        "step_seconds": 60,
+        "sections": {
+            "API": {
+                "Metrics": {
+                    metric: {
+                        "metric": metric,
+                        "unit": unit,
+                        "latest": 1.234567,
+                        "min": 1.234567,
+                        "max": 1.234567,
+                        "avg": 1.234567,
+                    }
+                    for metric, unit in (
+                        ("cpu_usage", "percent"),
+                        ("replica_count", "count"),
+                        ("p95_latency", "milliseconds"),
+                        ("responses_per_second", "responses_per_second"),
+                    )
+                }
+            }
+        },
+    }
 
     class FrozenDatetime(datetime):
         @classmethod
@@ -57,6 +81,8 @@ def test_metrics(monkeypatch, bounds, mode, deployment_metrics_response):
         "end_time": "2026-10-28T23:02:33+00:00",
         **options,
     }
+    if mode == "--summary":
+        params["summary"] = "true"
 
     def handler(request):
         assert request.method == "GET"
@@ -64,7 +90,9 @@ def test_metrics(monkeypatch, bounds, mode, deployment_metrics_response):
         assert dict(request.url.params) == params
         assert request.headers["x-tenant-id"] == "workspace-id"
         assert request.headers["x-api-key"] == "key"
-        return httpx.Response(200, json=charts)
+        return httpx.Response(
+            200, json=server_summary if mode == "--summary" else charts
+        )
 
     args = ["metrics", DEPLOYMENT_ID]
     for key, value in options.items():
@@ -73,19 +101,19 @@ def test_metrics(monkeypatch, bounds, mode, deployment_metrics_response):
         args.append(mode)
     result = invoke(monkeypatch, args, handler)
     assert result.exit_code == 0, result.output
-    expected = (
-        charts
-        if mode == "--raw"
-        else compact_deployment_metrics(
+    if mode == "--summary":
+        expected = server_summary
+    elif mode == "--raw":
+        expected = charts
+    else:
+        expected = compact_deployment_metrics(
             charts,
             DEPLOYMENT_ID,
-            start_time=params.get("start_time"),
-            end_time=params.get("end_time"),
-            summary=mode == "--summary",
+            start_time=params["start_time"],
+            end_time=params["end_time"],
         )
-    )
     assert json.loads(result.output) == expected
-    if mode != "--raw":
+    if mode is None:
         assert expected["start_time"] == params["start_time"]
         assert expected["end_time"] == params["end_time"]
     assert result.output.strip() == json.dumps(expected, separators=(",", ":"))

libs/cli/tests/unit_tests/test_deployment_metrics.py

@@ -53,22 +53,15 @@ def test_compact_metrics(deployment_metrics_response):
     assert result["sections"]["API"]["Latency"]["P95 Latency"]["unit"] is None
 
 
-def test_summary_and_requested_bounds(deployment_metrics_response):
+def test_requested_bounds(deployment_metrics_response):
     result = reshape(
         deployment_metrics_response,
         "id",
         start_time="2026-10-27T23:00:00Z",
         end_time="2026-10-28T01:00:00Z",
-        summary=True,
     )
     assert result["start_time"] == "2026-10-27T23:00:00Z"
     assert result["end_time"] == "2026-10-28T01:00:00Z"
-    for section in result["sections"].values():
-        if section == {"_empty": True}:
-            continue
-        for chart in section.values():
-            for series in chart.values():
-                assert set(series) == {"metric", "unit", "latest", "min", "max", "avg"}
 
 
 @pytest.mark.parametrize("values", [[], [None], [float("nan"), float("inf")]])