Compare patches, objective correctness, manual review, solution time, and known provider cost without treating availability or infrastructure failures as code quality.
Every route follows the same controlled path. A failure stays visible as evidence; it is not converted into a quality score or silently removed from the table.
hi request confirms the exact route.One attempt. No second chance. Retries cost time and money; the first delivered result is the one preserved and scored. Read the full workflow ↗
A focused JavaScript maintenance task executed through the clean-room OpenCode and Codex pipeline, with public tests, a private objective evaluator, immutable patches, time, and provider-reported cost.
The first broad patch comparison: 59 coding-agent candidates solving the same ledger task, with availability kept separate from quality scoring.
Diagnose and repair duplicate IDs, exact money parsing, sales, refunds, invalid rows, and daily ledger totals.
Implement feature overrides and deterministic percentage rollout from a detailed behavioral contract.
Extract one shared event-filtering helper while preserving the public API and all existing behavior.
Explore an unfamiliar module layout, locate the correct formatter, and repair account status labels.
Repair strict integer-range parsing and author meaningful tests that survive mutation testing.
Archived results from the older combined run. Audit found shared-history contamination, so these records are retained as evidence but are not part of the new clean-room nomination pages.
The public benchmark runner documents clean-room isolation, separate candidate testing and judging, objective evaluation, timing, cost policy, and immutable result artifacts.