ops-dashboard · cc-sonnet-5.5
claude-code · started 2026-10-08T14:45:10.591Z · horizon 420s · judge ai (openai:gpt-5.5), 5 frames scored
Area under curve
0.931
headline, 0-1
Tab title/icon
8.7s
browser chrome, not the page
First render
8.7s
anything on screen
First reviewable
32.7s
worth feedback
Final score
1.00
no regression
Correctness over time
Where the time went
100%
Model thinking 36.0sTool round trips 0.1s
| Phase | Time | Share | Side |
|---|---|---|---|
| Model thinking | 36.0s | 99.6% | agent |
| Tool round trips | 0.1s | 0.4% | agent |
| Dependency install | 0.0s | 0.0% | toolchain |
| Build | 0.0s | 0.0% | toolchain |
| Dev server boot | 0.0s | 0.0% | toolchain |
| First paint | 0.0s | 0.0% | toolchain |
| Unaccounted | 0.0s | 0.0% | unknown |
| Wall clock (cold start) | 36.2s | 100% |
Agent or toolchain
100%
Agent (model + tool round trips) 36.2s
Toolchain (install + build + boot + paint) 0.0s
Unaccounted 0.0s
Accounted for: 100.0% of wall clock. Agent reported 38.6s of API time; this run attributed 36.0s (delta 2.6s).
- Excludes 15.0s of deliberate idle observation (settle window); the curve still covers it.
Iteration: prompt to visible change
Follow-ups were injected into the same live session.
| Edit | First visible change | Correct change | Agent done | Tool calls | After agent | Broken for |
|---|---|---|---|---|---|---|
| Make the page heading blue. | 6.7s | 6.7s | 12.5s | 1 (Edit) | -5.8s | -- |
| Add an eleventh depot row for Porto: 6,480 shipments, 93.1% on-time, $19.95 cost per shipment. | 5.2s | 5.2s | 8.6s | 1 (Edit) | -3.4s | -- |
Judged frames
Only the heading and subtitle are visible; dashboard content is missing.
Only the styled header/subtitle are visible; required KPI, table, chart, and footer content are missing.
Visible header and KPI row are polished, but table, bars, and footer are absent from screenshot.
Strong dashboard visible, but bar chart and footer are missing from screenshot.
Strong dashboard implementation; only the required footer is not visible.
All requested dashboard elements are visible, accurate, aligned, and well designed.
Caveats
- 2 poll ticks were skipped because a capture overran the interval.
- judge: openai:gpt-5.5 does not accept a sampling temperature, so these frames were scored at the provider's default rather than at 0. Two runs of the same page can be scored differently, and that variance is inside the AUC. result.json records the temperature as null rather than as the value that was asked for, so p2p compare will not pool this run with one that was scored at a fixed temperature.