todo-app-scaffolded · pi-gpt-5.5
pi · started 2026-10-07T15:07:26.448Z · horizon 480s · judge ai (anthropic:claude-sonnet-5), 1 frames scored
Area under curve
0.947
headline, 0-1
First render
25.4s
anything on screen
First reviewable
25.4s
worth feedback
Final score
1.00
no regression
Correctness over time
Where the time went
96%
Model thinking 23.9sTool round trips 0.1sUnaccounted 1.0s
| Phase | Time | Share | Side |
|---|---|---|---|
| Model thinking | 23.9s | 95.7% | agent |
| Tool round trips | 0.1s | 0.3% | 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 | 4.0% | unknown |
| Wall clock (cold start) | 25.0s | 100% |
Agent or toolchain
96%
Agent (model + tool round trips) 24.0s
Toolchain (install + build + boot + paint) 0.0s
Unaccounted 1.0s
Accounted for: 96.0% 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 header background blue. | 5.2s | 5.2s | 7.3s | 1 (edit) | -- | -- |
| Add a fourth column called "Blocked" after Done. | 4.6s | 4.6s | 9.3s | 0 | -4.7s | -- |
Judged frames
Polished, fully styled task board matching brief with columns, cards, and add button.
Caveats
- 2 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.
- Iteration "header-blue" was corrected after the run: The brief's check read only background-color, and this header is blue through a linear-gradient, so it reported NEVER LANDED. Re-checked after the run with the corrected check (gradient colour stops count) against the run's final code: it passes. The page changed once after the prompt and held that state to the end of the window, so the edit landed at its first visible change.