Each challenge is one UI brief. Its inputs are frozen and versioned, so every output can be read against exactly what the tool was given.
An exact model and reasoning setting per Configuration and no subagents, but no shared sandbox: Codex used its own, and Claude Code and OpenCode had none until 2026-09-24. Runs are compared only within one protocol. How the protocols differ
Clear filtersNewest first
Design and build Fjora's responsive Taste the North campaign landing page from the frozen prompt, rules, facts, photography, and requirements.