feat(failures): show passing groups and an empty state

backend/src/ref_backend/api/routes/executions.py

@@ -87,6 +87,8 @@ async def list_recent_execution_groups(  # noqa: PLR0913, PLR0917
     offset: int = Query(0, ge=0),
     diagnostic_name_contains: str | None = None,
     provider_name_contains: str | None = None,
+    provider_slug: str | None = Query(None, description="Keep only the groups of this provider"),
+    diagnostic_slug: str | None = Query(None, description="Keep only the groups of this diagnostic"),
     dirty: bool | None = None,
     successful: bool | None = None,
     failure_reason: ExecutionFailureReason | None = None,
@@ -107,6 +109,7 @@ async def list_recent_execution_groups(  # noqa: PLR0913, PLR0917
     Supports filtering by:
     - diagnostic_name_contains (case-insensitive substring of the diagnostic slug)
     - provider_name_contains (case-insensitive substring of the provider slug)
+    - provider_slug and diagnostic_slug (exact slugs)
     - dirty
     - successful (filters by latest execution success)
     - failure_reason (filters by why the latest execution failed, e.g. `system_error`)
@@ -119,6 +122,8 @@ async def list_recent_execution_groups(  # noqa: PLR0913, PLR0917
     filters = ExecutionGroupFilter(
         diagnostic_contains=diagnostic_name_contains or None,
         provider_contains=provider_name_contains or None,
+        provider_slug=provider_slug,
+        diagnostic_slug=diagnostic_slug,
         include_superseded=False,
         include_retracted=False,
         dirty=dirty,

backend/tests/test_api/test_routes/test_executions.py

@@ -436,6 +436,24 @@ def test_execution_statistics_match_list(client: TestClient, settings) -> None:
     assert unsuccessful["total_count"] == listed["total_count"] - successful["total_count"]
 
 
+def test_execution_list_filters_by_exact_slugs(client: TestClient, settings) -> None:
+    """The slug filters match whole slugs, unlike the `_contains` filters."""
+    group = client.get(f"{settings.API_V1_STR}/executions?limit=1").json()["data"][0]
+    provider, diagnostic = group["provider_slug"], group["diagnostic_slug"]
+
+    matched = client.get(
+        f"{settings.API_V1_STR}/executions",
+        params={"provider_slug": provider, "diagnostic_slug": diagnostic, "limit": 1000},
+    ).json()
+    assert matched["total_count"] > 0
+    assert {(g["provider_slug"], g["diagnostic_slug"]) for g in matched["data"]} == {(provider, diagnostic)}
+
+    partial = client.get(
+        f"{settings.API_V1_STR}/executions", params={"diagnostic_slug": diagnostic[:-1], "limit": 1}
+    ).json()
+    assert all(g["diagnostic_slug"] == diagnostic[:-1] for g in partial["data"])
+
+
 def test_execution_list_pagination(client: TestClient, settings) -> None:
     page1 = client.get(f"{settings.API_V1_STR}/executions?limit=2&offset=0").json()
     page2 = client.get(f"{settings.API_V1_STR}/executions?limit=2&offset=2").json()

changelog/159.feature.md

@@ -0,0 +1 @@
+The failures page can now show passing execution groups, and lists models without failures with an empty state saying how many of their groups passed.

frontend/src/client/@tanstack/react-query.gen.ts

@@ -482,6 +482,7 @@ export const executionsListRecentExecutionGroupsQueryKey = (options?: Options<Ex
  * Supports filtering by:
  * - diagnostic_name_contains (case-insensitive substring of the diagnostic slug)
  * - provider_name_contains (case-insensitive substring of the provider slug)
+ * - provider_slug and diagnostic_slug (exact slugs)
  * - dirty
  * - successful (filters by latest execution success)
  * - failure_reason (filters by why the latest execution failed, e.g. `system_error`)
@@ -517,6 +518,7 @@ export const executionsListRecentExecutionGroupsInfiniteQueryKey = (options?: Op
  * Supports filtering by:
  * - diagnostic_name_contains (case-insensitive substring of the diagnostic slug)
  * - provider_name_contains (case-insensitive substring of the provider slug)
+ * - provider_slug and diagnostic_slug (exact slugs)
  * - dirty
  * - successful (filters by latest execution success)
  * - failure_reason (filters by why the latest execution failed, e.g. `system_error`)
@@ -684,7 +686,7 @@ export const executionsListMetricValuesQueryKey = (options: Options<ExecutionsLi
  *
  * Fetch metric values for a specific execution (both scalar and series)
  *
- * - `value_type`: Type of metric values - 'scalar', 'series', or 'all' (required)
+ * - `value_type`: Type of metric values - 'scalar' or 'series' (required)
  * - `format`: Return format - 'json' (default) or 'csv'
  * - `offset`: Number of items to skip (default 0)
  * - `limit`: Maximum number of items to return (default 50, max 500)
@@ -711,7 +713,7 @@ export const executionsListMetricValuesInfiniteQueryKey = (options: Options<Exec
  *
  * Fetch metric values for a specific execution (both scalar and series)
  *
- * - `value_type`: Type of metric values - 'scalar', 'series', or 'all' (required)
+ * - `value_type`: Type of metric values - 'scalar' or 'series' (required)
  * - `format`: Return format - 'json' (default) or 'csv'
  * - `offset`: Number of items to skip (default 0)
  * - `limit`: Maximum number of items to return (default 50, max 500)

frontend/src/client/sdk.gen.ts

@@ -151,6 +151,7 @@ export const executionsGetExecutionStatistics = <ThrowOnError extends boolean =
  * Supports filtering by:
  * - diagnostic_name_contains (case-insensitive substring of the diagnostic slug)
  * - provider_name_contains (case-insensitive substring of the provider slug)
+ * - provider_slug and diagnostic_slug (exact slugs)
  * - dirty
  * - successful (filters by latest execution success)
  * - failure_reason (filters by why the latest execution failed, e.g. `system_error`)
@@ -215,7 +216,7 @@ export const executionsListMetricValueFacets = <ThrowOnError extends boolean = f
  *
  * Fetch metric values for a specific execution (both scalar and series)
  *
- * - `value_type`: Type of metric values - 'scalar', 'series', or 'all' (required)
+ * - `value_type`: Type of metric values - 'scalar' or 'series' (required)
  * - `format`: Return format - 'json' (default) or 'csv'
  * - `offset`: Number of items to skip (default 0)
  * - `limit`: Maximum number of items to return (default 50, max 500)

frontend/src/client/types.gen.ts

@@ -2538,6 +2538,18 @@ export type ExecutionsListRecentExecutionGroupsData = {
          * Provider Name Contains
          */
         provider_name_contains?: string | null;
+        /**
+         * Provider Slug
+         *
+         * Keep only the groups of this provider
+         */
+        provider_slug?: string | null;
+        /**
+         * Diagnostic Slug
+         *
+         * Keep only the groups of this diagnostic
+         */
+        diagnostic_slug?: string | null;
         /**
          * Dirty
          */

frontend/src/components/failures/diagnosticFailuresCard.tsx

@@ -1,6 +1,6 @@
 import { Link } from "@tanstack/react-router";
 import { format } from "date-fns";
-import { ChevronRight, SquareArrowOutUpRight } from "lucide-react";
+import { ChevronRight } from "lucide-react";
 import { useState } from "react";
 import type { ExecutionGroup } from "@/client";
 import { Badge } from "@/components/ui/badge";
@@ -29,9 +29,12 @@ import {
   lastSucceededAt,
   reasonBreakdown,
 } from "@/lib/failures";
+import type { MipEra } from "@/lib/mipEras";
 import { FailureReasonBadge } from "./failureReasonBadge";
 import { HealthDot } from "./health";
 import { LogExcerpt } from "./logExcerpt";
+import { OpenGroupLink } from "./openGroupLink";
+import { PassingGroups } from "./passingGroups";
 
 const day = (date: string) => format(new Date(date), "d MMM yyyy");
 
@@ -91,15 +94,7 @@ function FailedGroupItem({ group }: { group: ExecutionGroup }) {
           <span className="hidden shrink-0 text-xs text-muted-foreground sm:inline">
             {format(new Date(execution.updated_at), "yyyy-MM-dd HH:mm")}
           </span>
-          <Link
-            to="/executions/$groupId/logs"
-            params={{ groupId: group.id.toString() }}
-            aria-label="Open execution group"
-            title="Open execution group"
-            className="shrink-0"
-          >
-            <SquareArrowOutUpRight className="size-4 text-blue-500 hover:text-blue-300 dark:text-blue-400 dark:hover:text-blue-300" />
-          </Link>
+          <OpenGroupLink groupId={group.id} />
         </div>
         <CollapsibleContent className="px-2 pb-2">
           <LogExcerpt groupId={group.id} executionId={execution.id} />
@@ -109,14 +104,20 @@ function FailedGroupItem({ group }: { group: ExecutionGroup }) {
   );
 }
 
-/** A diagnostic with failing execution groups, listing each with what its log says about the failure. */
+/**
+ * A diagnostic's failing execution groups, listing each with what its log says about the failure
+ *
+ * Given `passingEra`, it also lists the diagnostic's passing groups of that era.
+ */
 export function DiagnosticFailuresCard({
   failures,
   filters = {},
+  passingEra,
 }: {
   failures: DiagnosticFailures;
   /** What the failures were narrowed to, named in the summary */
   filters?: FailureFilters;
+  passingEra?: MipEra;
 }) {
   const [showAll, setShowAll] = useState(false);
   const { diagnostic } = failures;
@@ -179,6 +180,14 @@ export function DiagnosticFailuresCard({
               : `Show all ${groups.length} failed execution groups`}
           </Button>
         )}
+        {passingEra && failures.successful > 0 && (
+          <PassingGroups
+            diagnostic={diagnostic}
+            count={failures.successful}
+            mipEra={passingEra}
+            model={filters.model}
+          />
+        )}
       </CardContent>
     </Card>
   );

frontend/src/components/failures/failureFilters.tsx

@@ -1,6 +1,7 @@
 import { Check, ChevronsUpDown, X } from "lucide-react";
 import { useState } from "react";
 import { Button } from "@/components/ui/button";
+import { Checkbox } from "@/components/ui/checkbox";
 import {
   Command,
   CommandEmpty,
@@ -9,6 +10,7 @@ import {
   CommandItem,
   CommandList,
 } from "@/components/ui/command";
+import { Label } from "@/components/ui/label";
 import {
   Popover,
   PopoverContent,
@@ -21,7 +23,9 @@ import { cn } from "@/lib/utils";
 import { type DayRange, DayRangePicker } from "./dayRangePicker";
 
 const countText = ({ groups, diagnostics }: FailureCount) =>
-  `${groups} failed in ${diagnostics} ${diagnostics === 1 ? "diagnostic" : "diagnostics"}`;
+  groups === 0
+    ? "no failures"
+    : `${groups} failed in ${diagnostics} ${diagnostics === 1 ? "diagnostic" : "diagnostics"}`;
 
 /**
  * A searchable dropdown picking one value to filter failures by, or all of them
@@ -144,6 +148,8 @@ export function FailureFilterBar({
   allDays,
   failureDays,
   onDays,
+  passing,
+  onPassing,
   onClear,
   summary,
   disabled,
@@ -165,6 +171,8 @@ export function FailureFilterBar({
   /** Days with failures matching the other filters, marked in the calendar */
   failureDays?: Set<string>;
   onDays: (days: DayRange) => void;
+  passing: boolean;
+  onPassing: (passing: boolean) => void;
   onClear?: () => void;
   summary?: string;
   disabled?: boolean;
@@ -206,6 +214,17 @@ export function FailureFilterBar({
           onChange={onDays}
           disabled={disabled || !allDays}
         />
+        <div className="flex items-center gap-2 text-sm">
+          <Checkbox
+            id="failures-show-passing"
+            checked={passing}
+            onCheckedChange={(checked) => onPassing(checked === true)}
+            disabled={disabled}
+          />
+          <Label htmlFor="failures-show-passing" className="font-normal">
+            Show passing groups
+          </Label>
+        </div>
         {onClear && (
           <Button variant="ghost" size="sm" onClick={onClear}>
             <X className="size-4" />

frontend/src/components/failures/failuresEmptyState.tsx

@@ -0,0 +1,35 @@
+import { CircleCheck } from "lucide-react";
+import { Button } from "@/components/ui/button";
+
+/** Shown when nothing matching the filters is failing, with how many groups passed */
+export function FailuresEmptyState({
+  heading,
+  totals,
+  onShowPassing,
+}: {
+  heading: string;
+  /** The groups the filters cover, left out when a reason or time range narrows the failures */
+  totals?: { total: number; successful: number };
+  /** Offered when passing groups are hidden and some exist */
+  onShowPassing?: () => void;
+}) {
+  return (
+    <div className="flex flex-col items-center gap-3 rounded-lg border border-dashed py-8 text-center">
+      <CircleCheck className="size-8 text-green-600 dark:text-green-500" />
+      <div className="space-y-1">
+        <p className="font-medium">{heading}</p>
+        {totals && totals.total > 0 && (
+          <p className="text-sm text-muted-foreground">
+            {totals.successful} of {totals.total} execution groups passed at the
+            current version of each diagnostic.
+          </p>
+        )}
+      </div>
+      {onShowPassing && (
+        <Button variant="outline" size="sm" onClick={onShowPassing}>
+          Show passing groups
+        </Button>
+      )}
+    </div>
+  );
+}

frontend/src/components/failures/openGroupLink.tsx

@@ -0,0 +1,17 @@
+import { Link } from "@tanstack/react-router";
+import { SquareArrowOutUpRight } from "lucide-react";
+
+/** Opens an execution group's logs */
+export function OpenGroupLink({ groupId }: { groupId: number }) {
+  return (
+    <Link
+      to="/executions/$groupId/logs"
+      params={{ groupId: groupId.toString() }}
+      aria-label="Open execution group"
+      title="Open execution group"
+      className="shrink-0"
+    >
+      <SquareArrowOutUpRight className="size-4 text-blue-500 hover:text-blue-300 dark:text-blue-400 dark:hover:text-blue-300" />
+    </Link>
+  );
+}