Skip to content
evalship Get started →

evalship-hq / ponytail

Public

JS debounce | openrouter:anthropic/claude-opus-4.8

Ponytail vs caveman vs no-skill: same model, same tasks. Measures code LOC (deterministic) and token / JS debounce | openrouter:anthropic/claude-opus-4.8

Score on main

Oct 29: 6.00Oct 29: 1.00Oct 29: 1.00Oct 29: 1.00
Oct 29min 1.00, max 6.00Oct 29: 1.00

Last 4 runs

Stable across recent runs.

Recent results

CommitResultScoreDurationMetadata
7beb065 6 days ago Pass 1.00 3.6s provider: openrouter:anthropic/claude-opus-4.8, scoring: code_loc weight 0, assertions: ["javascript","javascript"], judge: deterministic
545983b 6 days ago Pass 1.00 2.9s provider: openrouter:anthropic/claude-opus-4.8, scoring: code_loc weight 0, assertions: ["javascript","javascript"], judge: deterministic
7e38f14 6 days ago Pass 1.00 3.0s provider: openrouter:anthropic/claude-opus-4.8, scoring: code_loc weight 0, assertions: ["javascript","javascript"], judge: deterministic
c01cd19 6 days ago Pass 6.00 3.9s provider: openrouter:anthropic/claude-opus-4.8, assertions: ["javascript","javascript"], judge: deterministic