fix(viewer): show scenario run model labels

tests/test_viewer_metrics.py

@@ -0,0 +1,81 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+import json
+import subprocess
+import textwrap
+import unittest
+from pathlib import Path
+from tempfile import TemporaryDirectory
+
+from tests.node_runner import node_supports_ts, node_ts_args
+
+
+ROOT = Path(__file__).resolve().parents[1]
+METRICS_SRC = ROOT / "viewer" / "src" / "lib" / "server" / "metrics.ts"
+
+
+@unittest.skipUnless(node_supports_ts(), "node binary lacks TypeScript support (need ≥ 22.6)")
+class ViewerMetricsTest(unittest.TestCase):
+    def _run_node(self, *, harness_dir: Path, script: str) -> subprocess.CompletedProcess[str]:
+        return subprocess.run(
+            ["node", *node_ts_args(), "--input-type=module"],
+            input=script,
+            text=True,
+            capture_output=True,
+            cwd=harness_dir,
+            check=False,
+        )
+
+    def test_audit_metrics_fall_back_to_tester_model_for_target_label(self) -> None:
+        with TemporaryDirectory(dir=ROOT / "viewer") as tmp_dir:
+            harness_dir = Path(tmp_dir)
+            source = METRICS_SRC.read_text(encoding="utf-8")
+            source = source.replace("from '$lib/judgment.js';", "from './judgment.js';")
+            source = source.replace("from './dimensions.js';", "from './dimensions.js';")
+            (harness_dir / "metrics.ts").write_text(source, encoding="utf-8")
+            (harness_dir / "judgment.js").write_text(
+                textwrap.dedent(
+                    """\
+                    export function getRequiredBaseMetricNames() { return ['policy_violation', 'overrefusal']; }
+                    export function isBooleanFlag(value) { return typeof value === 'boolean'; }
+                    export function getRecordFlag(record, name) {
+                      const value = record?.verdict?.dimensions?.[name];
+                      return typeof value === 'boolean' ? value : null;
+                    }
+                    export function isSuccessfulJudgment(record) { return record?.judge_status !== 'error'; }
+                    """
+                ),
+                encoding="utf-8",
+            )
+            (harness_dir / "dimensions.js").write_text(
+                "export function loadDimensions() { return []; }\n",
+                encoding="utf-8",
+            )
+
+            script = textwrap.dedent(
+                """\
+                const { computeAuditRunMetrics } = await import('./metrics.ts');
+                const metrics = computeAuditRunMetrics([{
+                  test_case_id: 'scenario-1',
+                  behavior: 'medical advice',
+                  tester_model: 'azure/gpt-5.4-1',
+                  judge_model: 'azure/gpt-5.4-1',
+                  judge_status: 'ok',
+                  verdict: { dimensions: { policy_violation: false, overrefusal: false } },
+                  metadata: { turns_count: 2, stop_reason: 'max_turns' }
+                }]);
+                console.log(JSON.stringify(metrics));
+                """
+            )
+            result = self._run_node(harness_dir=harness_dir, script=script)
+
+            self.assertEqual(result.returncode, 0, msg=f"{result.stdout}\n{result.stderr}")
+            payload = json.loads(result.stdout)
+            self.assertEqual(payload["target"], "azure/gpt-5.4-1")
+            self.assertEqual(payload["tester_model"], "azure/gpt-5.4-1")
+            self.assertEqual(payload["judge_model"], "azure/gpt-5.4-1")
+
+
+if __name__ == "__main__":
+    unittest.main()

viewer/src/lib/server/metrics.ts

@@ -113,7 +113,7 @@ export function computeAuditRunMetrics(scores: AuditScore[]): AuditRunMetrics |
 		policy_violation_rate: dimensionRate(dimensions, 'policy_violation'),
 		overrefusal_rate: dimensionRate(dimensions, 'overrefusal'),
 		dimensions,
-		target: scores[0]?.target ?? '',
+		target: scores[0]?.target ?? scores[0]?.tester_model ?? '',
 		tester_model: scores[0]?.tester_model ?? '',
 		judge_model: scores[0]?.judge_model ?? ''
 	};

viewer/src/routes/suite/[suite_id]/[run_id]/+page.svelte

@@ -760,9 +760,9 @@
 				{/if}
 			</div>
 			<div class="mt-2 text-sm text-text-muted">
-				Evaluation target: <span class="font-mono text-text">{data.metrics?.target ?? data.auditMetrics?.target ?? '—'}</span>
+				Evaluation target: <span class="font-mono text-text">{data.metrics?.target || data.auditMetrics?.target || '—'}</span>
 				<span class="mx-2 text-text-muted/50">·</span>
-				Judge: <span class="font-mono text-text">{data.metrics?.judge_model ?? data.auditMetrics?.judge_model ?? '—'}</span>
+				Judge: <span class="font-mono text-text">{data.metrics?.judge_model || data.auditMetrics?.judge_model || '—'}</span>
 			</div>
 			{#if data.manifest?.stages}
 				<button