Tracer
← overview

polyphony

9 signals · last run 11h ago · RTrentJones/Polyphony

✗ failingbadgeJSON
SignalKindLatest modelTrendLatestFreshnessCommit
eval · cel · ciqualityclaude-opus-5-52026-10-23 22:51 UTC · 0% · fail2026-10-24 00:29 UTC · 0% · fail2026-10-24 05:47 UTC · 100% · pass✓ 100%3d agob46d695
eval · dracula · ciqualitygemini-3.5-flash-lite2026-07-13 04:36 UTC · 100% · pass2026-07-13 05:21 UTC · 100% · pass2026-07-13 09:52 UTC · 86% · fail2026-07-13 15:26 UTC · 57% · fail2026-07-14 02:32 UTC · 67% · fail2026-07-14 07:24 UTC · 86% · fail2026-07-14 20:01 UTC · 100% · pass2026-07-15 07:32 UTC · 50% · fail2026-07-16 07:27 UTC · 71% · fail2026-07-20 09:40 UTC · 83% · fail2026-07-27 10:22 UTC · 25% · fail2026-08-03 10:10 UTC · 60% · fail2026-08-10 08:14 UTC · 50% · fail2026-08-17 07:21 UTC · 60% · fail2026-08-24 07:37 UTC · 33% · fail2026-08-31 14:03 UTC · 50% · fail2026-09-07 12:44 UTC · 50% · fail2026-09-14 12:59 UTC · 50% · fail2026-09-21 12:58 UTC · 50% · fail2026-09-28 14:11 UTC · 50% · fail2026-10-05 14:56 UTC · 50% · fail2026-10-12 15:43 UTC · 50% · fail2026-10-19 14:00 UTC · 75% · fail2026-10-23 22:22 UTC · 86% · fail2026-10-24 00:14 UTC · 100% · pass2026-10-24 05:21 UTC · 86% · fail2026-10-24 05:58 UTC · 100% · pass2026-10-24 06:14 UTC · 100% · pass2026-10-24 09:36 UTC · 100% · pass2026-10-25 00:24 UTC · 100% · pass2026-10-25 04:45 UTC · 100% · pass2026-10-26 05:16 UTC · 100% · pass2026-10-26 10:06 UTC · 0% · fail✗ 0%11h ago90363f2
eval · frankenstein · ciqualitygemini-3.5-flash-lite2026-07-13 04:36 UTC · 100% · pass2026-07-14 02:33 UTC · 67% · fail2026-07-20 09:42 UTC · 67% · fail2026-07-27 10:29 UTC · 20% · fail2026-08-03 10:17 UTC · 40% · fail2026-08-10 08:20 UTC · 50% · fail2026-08-17 07:32 UTC · 25% · fail2026-08-24 07:43 UTC · 50% · fail2026-08-31 14:13 UTC · 33% · fail2026-09-07 12:54 UTC · 33% · fail2026-09-14 13:09 UTC · 33% · fail2026-09-21 13:09 UTC · 33% · fail2026-09-28 14:22 UTC · 33% · fail2026-10-05 15:08 UTC · 33% · fail2026-10-12 15:49 UTC · 50% · fail2026-10-19 14:58 UTC · 100% · pass2026-10-23 08:18 UTC · 86% · fail2026-10-23 22:47 UTC · 71% · fail2026-10-24 00:24 UTC · 33% · fail2026-10-24 05:42 UTC · 71% · fail2026-10-24 06:05 UTC · 100% · pass2026-10-24 06:21 UTC · 100% · pass2026-10-24 07:36 UTC · 0% · fail2026-10-24 09:49 UTC · 100% · pass2026-10-25 00:37 UTC · 100% · pass2026-10-25 04:55 UTC · 100% · pass2026-10-26 05:25 UTC · 0% · fail2026-10-26 10:00 UTC · 0% · fail✗ 0%12h ago90363f2
eval · wuthering · ciqualitygemini-3.5-flash-lite2026-10-26 10:20 UTC · 0% · fail✗ 0%11h ago90363f2
5 stale signals (no recent run; not counted in health)
SignalKindLatest modelTrendLatestFreshnessCommit
eval · ci-eval-backoff-celqualitygemini-2.5-flash2026-10-12 20:58 UTC · 0% · fail✗ 0%◷ 14d ago—
eval · ci-eval-diag-celqualitygemini-2.5-flash2026-10-12 20:28 UTC · 0% · fail✗ 0%◷ 14d ago—
eval · ci-judge-fix-celqualitygemini-2.5-flash2026-10-13 07:39 UTC · 100% · pass✓ 100%◷ 14d ago—
eval · ci-post-reset-celqualitygemini-2.5-flash2026-10-13 07:22 UTC · 0% · fail✗ 0%◷ 14d ago—
eval · ci-spec-merge-celqualitygemini-2.5-flash2026-10-12 20:05 UTC · 0% · fail✗ 0%◷ 14d ago—

eval · dracula · ci (33 recent runs · all runs · compare models)

Pass rate per run

Metrics

Numeric results reported with each run, one chart per metric so each keeps its own scale.

attribution.accuracy0.600
attribution.chance0.333
attribution.score0.600
continuity.detection_recall1
continuity.false_positive_rate0.500
continuity.score1
extraction.f10.609
extraction.precision0.438
extraction.recall1
extraction.score0.609
ingestion.score0.647
outline.score1

Cases across the last 15 runs (oldest → latest ▾)

8 cases · ⚠ 1 flaky
case▾pass
⚠ eval:extraction✗✗✗✗✓✗✓✗✓✓·····40%
eval:ingestion✗✗✗✗✓✓✓✓✓✓·····60%
eval:outline····✗✓✓✓·······75%
eval:attribution·····0.53·0.630.470.60·····reported
eval:continuity✓✓✓✓·✓✓✓✓✓·····100%
eval:prose·····✓✓✓··✓✓✓✓·100%
eval:retrieval✓✓✓✓✓✓✓✓✓✓·····100%
eval:revise·····✓·✓·······100%