ops-dashboard · pi-gemini-3.8-flash
pi · started 2026-10-07T09:43:21.165Z · horizon 420s · judge ai (anthropic:claude-sonnet-5), 2 frames scored
Area under curve
0.893
headline, 0-1
Tab title/icon
45.1s
browser chrome, not the page
First render
45.1s
anything on screen
First reviewable
45.1s
worth feedback
Final score
1.00
no regression
Correctness over time
Where the time went
99%
Model thinking 1m 52sTool round trips 0.0sUnaccounted 1.0s
| Phase | Time | Share | Side |
|---|---|---|---|
| Model thinking | 1m 52s | 99.1% | agent |
| Tool round trips | 0.0s | 0.0% | 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 | 1.0s | 0.9% | unknown |
| Wall clock (cold start) | 1m 53s | 100% |
Agent or toolchain
99%
Agent (model + tool round trips) 1m 52s
Toolchain (install + build + boot + paint) 0.0s
Unaccounted 1.0s
Accounted for: 99.1% of wall clock.
- 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. | 6.6s | 6.6s | 10.9s | 1 (edit) | -4.3s | -- |
| Add an eleventh depot row for Porto: 6,480 shipments, 93.1% on-time, $19.95 cost per shipment. | 15.0s | 15.0s | 22.4s | 4 (read, edit) | -7.3s | -- |
Judged frames
All criteria met: complete, accurate, well-designed dashboard matching brief precisely.
All criteria met: polished, accurate, well-structured operations dashboard.
Caveats
- 1 poll ticks were skipped because a capture overran the interval.
- judge: anthropic:claude-sonnet-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.